Skip to main content

Full text of "dec :: Books :: ORourke Writing for the Reader 1976"

See other formats


* m 


by John O'Rourke 


Digital Software Publications 

prepared by 
Digital Software Document Preparation 

digital equipment corporation 
maynard. massachusetbs 

Copyright © 1976 by Digital Equipment Corporation 




A Verbal Communication System; The 
Sender; The Means of Transmission; 
The Receiver; Noise 


Gunning's Warning to Technical Writers; 
The Gulf Between Writers and Readers; 
Wrong Attitude; Right Attitude; The 
Writer's Purpose; Other Faults Noted by 
the Experts ; Conclusion 


One Reader Profile; Another Reader Profile; 
Writer's Grammar versus Reader's Grammar; 
Factors Impeding Comprehension; Long Sen- 
tences and Long Words; Sentence Structure; 
Position of Information in a Paragraph; Posi- 
tion of Adverbial Clauses in a Sentence; One 
Clause Instead of Two; Changes in the Logi- 
cal Order of Your Exposition; Subject Com- 
plements Instead of Object Complements; 
Factors Increasing Comprehension; Principle 
of Least Effort; Understanding and Misunder- 


Unity; Coherence; The Haystack Paragraph; 
The Ideal System 


Language and Reality; The Ladder of 
Abstraction; Abstract Words vs. Concrete 
Words; What's Good About Abstract 
Words? What's Wrong With Abstract 
Words? Making Abstract Sentences More 
Meaningful; Summary 



Required Memory Level; Sentence Length; 
Some Suggestions for Reducing Complexity 


Recognizing the Passive; The Case 
Against the Passive; The Case for 
the Passive; The Passive in Literature 
and Technical Writing; Shifting from 
the Passive to the Active 


What is Density? Adjective Strings; 
Stuffed Paragraphs; Unexplained 
Series; Lumpy Paragraphs; Summary 
of Effects and Remedies 


WORDS . . . 9-1 

Jargon; Big Words; Deadwood; Summary 

REFERENCES . . . . ......... R-l 

Articles, Dissertations and Monographs; 
Books (Recommended Reading) 


You should answer this question before you turn to 
Chapter 1. 

This book is for you if you are, first of all, a person 
who has a firm grasp of a technical subject but is rela- 
tively weak in the principles of written communi- 
cation. That is not to say that you have no back- 
ground in English. Rather, you've had the usual 
high school English courses, with perhaps two addi- 
tional courses in college or technical school. So 
you know clauses and sentences and parts of speech. 
And you either have at your fingertips the accepted 
rules of punctuation and usage or know where to 
get such information when you need it. 

Too, you are no stranger to writing: you've written 
letters and memos and reports. But you are new to 
this task of writing technical manuals for a large 
and varied readership. So you need a book of tips 
on how best to write for that body of readers. You 
need a book of ideas and principles. On the other 
hand, you don't want to be burdened with rules to 
memorize and exercises to do. 

If you fit this reader profile, then WRITING FOR THE 
READER is indeed for you. 


On the West Coast they tell the story of a plumber who 
started using hydrochloric acid on clogged pipes. Though 
he was pleased with the results, he wondered if he could 
be doing something wrong. So he wrote to Washington 
to get expert advice on the matter. In six weeks he 
received the following reply: 

"The efficacy of hydrochloric acid in the subject 
situation is incontrovertible, but its corrosiveness 
is incompatible with the integrity of metallic 

The plumber, who was short on formal education but 
long on hope, was elated. He shot a thank-you letter 
back to Washington. He told them he would lose no time 
in informing other plumbers about his discovery. Five 
weeks later he got another message: 

"In no case can we be presumed responsible for 
the generation of pernicious residues from hydro- 
chloric acid, and we strongiy recommend, there- 
fore, than an alternative method be utilized." 

The plumber was delighted. He sent his third letter in 
the next mail. In it, he said that about 15 plumbers in 
his city were now using hydrochloric acid for pipes. All 
of them liked it. Now he wondered whether the good 
people in Washington could help him spread the news of 
his discovery to plumbers throughout the country. At 
this point, the correspondence fell into the hands of a 
rare Washington bureaucrat - one who knew how to 
write to plumbers. Within a week the plumber was 
reading these words: 

"Stop using hydrochloric acid. And tell your 
friends to stop too. It eats the hell out of pipes." 

Certainly the letters in this interchange are a far cry 
from technical writing. Nevertheless, they offer a lesson 


to the new technical writer: Write so your reader can 

The chief function of a technical writer is to inform - to 
supply the reader with technical facts. Thus, each 
sentence you write must be aimed at the reader. If your 
sentences are not so aimed, you fail as a technical writer. 
Your message is like an arrow shot blindly into the air. It 
will fall somewhere. But the chances are that it will not 
hit the target. 

In other words, you, as a technical writer, cannot 
concern yourself merely with putting words on paper, as 
does the writer of a diary or the ill-fed poet in a garret. 
These people can enjoy the luxury of self-expression. 
They can exist without readers. Or they can, if they 
wish, write to parade their knowledge before the reader 
rather than to convey it to him. But you must 
concentrate on how best to convey meaning to the 
reader; you must strive to make your sentences mean the 
same to him as they do to you. 

It's too bad, in a way, that the term technical writer is 
used at all. The writer part of it has lost most of its 
meaning. Doesn't everyone write? The answer to that 
question is all around you. Programmers write, managers 
write, marketing people write, secretaries write. Look at 
the mountain of memos and reports these people 
produce. As each day brings a stream of such papers 
across your desk, you're forced to strain eye and mind 
to discover meaning amid the welter of words. Sure, 
everyone can write. But daily experience tells you only a 
few can communicate. 

In this book we will often use the term communicator 
instead of writer. In doing so, we realize we're replacing 
a word of two syllables with one of five, but we think 
the longer term is more descriptive of the unique skill of 
the technical writer. Consider for a moment the origin of 
the word communicator. It comes from the Latin 
communis, which means common. And as we've already 
emphasized, the main business of the technical writer is 
to communicate: to pass technical information along to 
the reader so that it becomes the common property of 


The remainder of this chapter will continue the emphasis 
on communication as the key to the whole art of 
technical writing. First you will see a verbal communi- 
cation system in operation. Then you will get a closeup 
of each element of that system. 


Figure 1 is a simple diagram of a communication system, 
as seen in communication theory. 



Means of transmission 

Figure 1 . A Verbal Communication System 

Each system has a sender, a receiver, and a means of 
transmission. The sender is the writer; the receiver is the 
reader. The means of transmission, represented by the 
broken line in the diagram, is in our case written 
language. The wavy line stands for noise. Noise is 
anything in the system that changes or distorts the 
meaning being transmitted. In all transmission, you the 
communicator try constantly to eliminate noise. 

Since clear, accurate meaning is what the communicator 
always strives to transmit, you can get some idea of the 
difficulty of the task by looking at Figure 2. It shows 
the four forms of meaning operating in the communi- 
cation of technical information. 








Figure 2. Four Forms of Meaning 


The source is the meaning in specifications, flow-charts, 
schematics, oral and written reports, and any other 
documents you use to get the facts for your writing. 
This meaning will always be different from the meaning 
(2) that lodges in your head after reading those 
documents. The degree of difference depends on your 
skill in interpreting the language of the source. Notice in 
Figure 2 that the size of 1 and 2 is roughly equivalent. 
The difference occurs mainly around the edges. This 
happens whenever words are used in the source, for 
words are variable in meaning. Thus, you will interpret 
some words right and others wrong. But the greater your 
skill in decoding the language of the various documents, 
the more the data in your head will approximate that of 
the source. 

You further distort and lose meaning when you start to 
write. For one thing, you aim to write only what the 
reader needs to know. Thus, you must select certain 
facts from the array in your head and reject others. 
Secondly, since most of the words you choose have 
multiple meanings, you always send additional - if 
unintentional - meaning along with your technical facts. 
For these reasons, 3 in the diagram is different from 2. 
Happily, in this instance it is only slightly different. 

Finally, the receiver, or reader, extracts a different 
meaning (4) from your written language, because it is 
impossible for all the words to mean the same to him as 
they do to you. Thus the meaning that ultimately resides 
in the reader's head is different in various respects from . 
the other three meanings. 

All these things considered, we can rightly call the 
communication in Figure 2 effective because the reader's 
head contains a sizable chunk of the meaning originally 
in the source. You can, however, further improve the 
communication. Whenever you get feedback from the 
reader, you modify the wording of the language (3). 

But the transmission can never be perfect. That is what 
makes your task as a communicator continue long after 
the software has gone to the customer. You have to 
struggle to make 3 clear the first time, and then you 
strive just as hard for clearness each time you modify 3 
because of feedback from the reader. In this way you 


avoid the defective communication of the secretaries, 
managers, etc. mentioned at the beginning of this 
chapter. A diagram of their faulty communication 
appears below (Figure 3). 



Figure 3. Faulty Communication of Ideas 


The blunt truth about the job of communication is that 
the entire burden of it rests on the writer's shoulders. 
You are the only one to blame if the process fails. 
Hence, yours is a vital role. And here is a partial list of 
what is expected of you as the sender in the communi- 
cation system. (You will get a detailed view of yourself 

1 . You are totally responsible for getting clear, 
accurate information from the source. 

2. You must gather the data on the experience 
and background of the reader so that you 
will use terms and other symbols the reader 
will understand. 

3. You're expected to have a firm grasp of the 
needs of the reader so that you will know 
which facts to communicate. 

4. You must be motivated by the desire to 
make the reader understand. This feeling 
guides you in organizing the technical infor- 
mation and in emphasizing important points. 

5. You are presumed to have mastered the art 
of communicating: making your words mean 


what you want them to mean in the mind of 
the reader. 

6. You must realize that language does not 
have to be elegant. You have to see it 
correctly as a tool for transferring ideas. 

7. You must constantly monitor your docu- 
ments to ensure that self-expression does not 
take the place of informative writing. 

8. You must remember that the deeper the 
subject, the harder you must work to make 
it understandable. 

9. You must never lose sight of your overall 
goal: to simplify the complex for the reader. 


The written language of the verbal communication 
system is called exposition. It is the link between the 
outside world and the ideas in your mind. Since its 
purpose is to inform, it appeals to the reader's intellect, 
not to his feelings or imagination. Ideally, you should 
make it clear and unambiguous, sending ideas by the 
shortest, most efficient route. It should always give the 
reader all he needs to know about the subject. Above all, 
you should write it oh a level that the reader can 
understand. Remember the Washington experts who 
didn't know how to write to the plumber. 


The receiver of your exposition is the reader. He comes 
to your writing to learn some technical facts. That 
learning may be a knowledge of ideas and their 
relationships. Or it may be a grasp of procedure- 
knowledge of the steps to follow to do a certain task. In 
either event, communication takes place when he ex- 
tracts from your writing the meaning you put into it; 
that is, when his interpretation of the meaning closely 
approaches yours. Only then can he perform the 
procedure you wrote about, or discuss the ideas you 

However, if he cannot decode your message correctly, 
communication breaks down. And poor communication 
is mostly a matter of interpretation. Sometimes the 
reader responds more to the implied meanings (conno- 
tation) of the words than to their literal meanings 
(denotation). Sometimes he distorts the meaning be- 
cause the language is too difficult for him. Most of the 


time he goes astray because of the various kinds of noise 
accompanying the message. 


In communication theory noise is any unnecessary or 
misleading information conveyed by your writing. Like 
static or other interference on radio or television, noise 
impedes communication. It prevents the reader from 
getting the meaning intended. 

The reader experiences noise as a lack of clearness. Such 
unclearness can range from slight trouble in under- 
standing a word or phrase, to total inability to under- 
stand any of the message. A lot of noise, or unclearness, 
causes the loss of meaning shown in Figure 3. 

Noise can originate in the writer's failure to extract 
clear, accurate information from his source. Most of the 
time, however, noise is caused by the writer's style. It 
arises from the six major ills of technical writing 
discussed in Chapters 4 through 9 of this book: 

1 . Poor organization of paragraphs. 

2. Excessive use of abstract words. 

3. The fog of sentence complexity. 

4. Overuse of the passive voice. 

5. Bunching together of too many technical 

6. Vague, empty, and difficult words. 





All exposition, according to the experts, is written on 
two different levels: the level of literacy and the level of 
competence. Literacy has to do with the so-called 
mechanics of writing — with punctuation, usage, and 
grammar, for example. (And you'll note that the 
mechanics lend themselves to testing and measurement.) 

Do you know what a sentence is? If so, you have 
mastered one phase of literacy. Can you tell nouns from 
verbs and prepositions? Can you use periods and 
question marks correctly? Do you know the difference 
between principle and principal 1 * Do you start each 
sentence with a capital letter? Can you detect and 
correct a run-on sentence? Are you a good speller? 

Such questions test your knowledge on the literacy level. 
But answering them all in the affirmative does not make 
you a good writer. Regardless of the emphasis that 
schools nowadays put on literacy, it is not the really 
important level of writing. In fact, at least one authority 
implies that literacy is not a part of good writing at all. 

The important level of writing involves the ability to 
move ideas from your head to the reader's. This level is 
what Edwin R. Clapp of Western Washington State 
College calls the level of competence. 

Competence in writing, then, has to do with how well 
you can pass technical facts along to the reader. Unlike 
literacy, it doesn't lend itself to objective tests or 

Can you recognize a topic and then explore it to 
discover its limits? If so, you have some skill on the 
competence level. Can you tell which ideas belong to 
that subject? And which do not belong? Can you write 
those ideas in the same order as they appear in your 


thinking? In other words, can you show one idea as the 
effect of the other? Is one the whole and the other a 
part? Can you show this relationship in writing? Does 
one follow the other? How do you express that 
sequence? Can you similarly express likenesses and 
differences among your ideas? How about equality and 
subordination? Are you proficient in writing definitions 
and classifications of your ideas? Can you develop 
theories or laws or principles from them? 

These questions show that competence in writing is 
connected with competence in thinking. In fact, they 
cannot be separated. Thus a writer can never be both 
skillful on the competence level of writing and fuzzy in 
his thinking. He can, however, be very skillful on the 
literacy level of writing and very incompetent in his 

Notice that the experts mentioned in the following 
sections are concerned only with competence. They 
don't stoop to deal with literacy at all. Rather, they harp 
on clearness, organization, exactness, relevance, and 
effectiveness. These are the marks of competence in 


About thirty years ago articles in the popular magazines 
were dry statements of fact backed up by cold statistics. 
The style was consistently stiff and dull, and the 
competence of the writing was low. So readers, finding 
them hard to read, sent loud complaints to the maga- 
zines. Not only did the readers voice their dislike of the 
articles, but they also specified what they wanted in 
place of them. The result is the modern article, which in 
contrast with the old is highly competent because it is 
highly readable. 

The point is that magazine articles changed because the 
magazines themselves responded to reader dis- 

Invariably, this is what happens when publications must 
depend on the reader for their existence. So says Robert 
Gunning in his book THE TECHNIQUE OF CLEAR 
WRITING. Either they produce what the reader wants, 
or they go out of business because the reader goes 


Such, in Gunning's opinion, is not the case with 
technical writing. The technical audience cannot simi- 
larly make its demands felt because it cannot go 
elsewhere for its technical information. Thus, the tech- 
nical writer can afford to write in a vacuum. But this 
situation will not last forever. Eventually, the demand 
for more readable technical writing will be as effective as 
was the demand for more readable magazine articles. 


When talking about writers, Gunning pulls no punches. 
He boldly asserts that the writer has created a wide gulf 
between himself and the reader - a gulf impeding the 
communication of ideas. And he goes on to describe the 
distressing situation. 

"The gulf between writers and readers is very 
great. Readers want careful organization in what 
they read. They desire concreteness to help them 
picture and apply ideas. They like variety, it 
maintains their interest. They prefer short but 
variable sentences and not too rich a mixture of 
hard words. 

"Writers, on the other hand, enjoy ^//-expression. 
They would as lief use abstractions to which they 
give their own special meanings. And a writing job 
goes easier and faster if you can simply set down 
the facts without the exacting thought needed for 
careful organization." 

Here, Gunning is speaking about writers in general. But 
another author singles out the technical writer for his 
part in creating the gulf. In ANALYTICAL WRITING, 
Thomas Johnson calls the technical writer to task for (1) 
lack of organization, (2) impersonal style, (3) use of big 
words, and (4) reluctance to state an idea directly. 


Johnson implies that the writer creates the gulf delib- 
erately. A better explanation blames the writer's attitude 
— his view of the writing task. As you will see later, 
Gunning supports this view. 

The new writer, according to the better explanation, is 
apt to have a wrong attitude toward writing. Talking 
about technical matters involves him in no difficulty at 


all. In the role of speaker he is usually logical and 
straightforward. But when he sits down to write, he 
changes. He is not himself. His natural directness 
disappears; he becomes stiff and formal, and, striking a 
pose, proceeds to erect a wall of artificiality between 
himself and the reader. Whereas in speaking he would 

"You must learn the operating system before you 
start working as a data base administrator." 

In writing it becomes: 

"Familiarity with the operating system must be 
acquired prior to attempting the undertaking of 
the role of Data Base Administrator." (DEC 

What transforms a clear speaker into a murky writer? 
The experts offer a number of reasons. Robert Gunning, 
for one, has strong feelings about the matter: 

"Many writers who are set the job of writing for a 
technical group believe their first task is to 
lengthen sentences and increase the mixture of 
polysyllables. They read technical communications 
in the new field, noting all odd departures from 
standard English. Thus in addition to picking up 
the technical vocabulary they are after, they 
absorb the bad habits of written expression that 
have impeded communication for years." 

Another view is that the new writer wants to impress the 
reader with his knowledge of the English language. He 
chooses to forget that writing and speaking have the 
same goal: to move ideas from one head to another. So 
he abandons the level of competence he used to good 
effect in speaking and descends to the level of literacy. 

A third body of opinion says that the new writer is stiff 
and formal because he wants to play safe. Observing the 
style of other technical documents, he fears his will seem 
unprofessional if written differently. 

A final view is that his writing is complex because his 
thinking is complex. He refuses to take the trouble to 


simplify it for the reader. In other words, he refuses to 
work for competence in his writing. 


Well, that's the ailment. What's the cure? The experts 
offer the following suggestions: 

1. Always remember that your chief concern is 
to get your content across. 

2. Often ask yourself "What am I doing?" Be 
honest in your answer. If your answer is that 
you are trying to impress the reader, then in 
the interest of competence you must change. 

3. Put your voice in your writing. This is 
Gunning's famous dictum. He says that 
when you talk you use your own voice, 
which is as uniquely you as your finger- 
prints. When you write, however, you forget 
your voice, with the result that your writing 
is not you. Competent writing, he says, has 
"voice" in it; if you know the writer, you 
can easily imagine him speaking to you as 
you read his writing. 

4. Copy the good points of conversational 
style. Use short sentences and short words. 
Divide ideas into small pieces and give the 
reader one or two pieces per sentence. Avoid 
inverted sentences. (Here is an example of an 
inverted structure that does not normally 
appear in conversation: "Sure he was that it 
wouldn't happen." You will be more effec- 
tive if you change it to the normal subject- 
verb-complement format of conversation: 
"He was sure that it wouldn't happen.") 

5. Use a 4-step approach to writing in order to 
retain "voice" and clarify thought. First, 
write as quickly as possible, as the ideas 
come to you. (Don't revise at all at this 
point.) Second, analyze your written expres- 
sion to uncover errors or gaps in your 
thinking. Third, revise what you have 
written to make it correct, clear, and exact. 
Finally, read it aloud to yourself to make 
sure your voice is in it. 


Note that these suggestions help you to acquire com- 
petence in writing. They help you transfer ideas from 
your head to the reader's. 


A lot of writing misses the mark, experts say, because 
the writer takes his eye off the target. In other words, he 
forgets his purpose. 

You should have the purpose of your document clear in 
your mind before you begin to write. Write it down in a 
single sentence and post it where you can look at it 
often. For example, you may write it as follows: "I want 
to explain all the duties of a data base administrator to a 
manager." Or, "I intend to describe all the components 
of this operating system so that a manager will under- 
stand what they are and what they do." Or, "I will 
explain SNOBOL so that the reader will be able to write 
a practical program after he finishes reading the 

All these purposes have to do with enlarging the reader's 
knowledge or giving him a new skill. As such, they are 
your chief concern as a technical writer. But you should 
also be aware of other purposes. You have to consider 
your manual in relation to those of competitors. Is a 
prospective customer likely to turn to a competitor 
because your manual does not explain DEC's software as 
well as the competitor's manual does his? Too, are you 
interested in gaining the largest possible readership for 
your document? 

If either of these purposes applies in your case, then 
you'd better aim to make your document highly 
readable. This is the suggestion of George Klare, author 
Not only will highly readable writing get the reader to 
accept your manual over another, but it will also attract 
more readers. He says that readers like readable writing 
because they can read it faster and get more out of it. 


Most of the foregoing statements are pretty general. But 
here are some specific faults the experts see as harmful 
to technical communication. Again, they all concern 


competence in writing or, rather, the lack of it. For each 
fault, we have included at least one example from a DEC 

1 . The writing is dull and dry because words do 
not say what they mean and syntax gives 
false cues that force the reader to go back 
and read again. "Reading in that case," says 
Jacques Barzun and Henry Graff "is like 
wading through a swamp." 

(Example) "All files are subject to being 
compressed with the exception of installed 
task files, the checkpoint file, and the save 
image file. Because of system linkages, these 
tasks (which are already contiguous) are left 
in their original position." 

(Comment) To what does the phrase these 
tasks refer? To the task files? Or to all three 
files? The reader has to reread because 
reference is unclear. 

(Example) "All privacy locks and keys are 
assigned by the Data Base Administrator, 
and the strictest security must surround 
their use. They should not, for instance, 
appear in the DBD or any other document 
which may be viewed by more than one user 
or multiple personnel." 

(Comment) What does multiple personnel 
mearil Is it different from more than one 
userl The reader has to reread looking for 
clues to the meaning. Unfortunately, there 
are no clues in the entire paragraph. 

(Example) "This manual describes DBMS-10 
from the point of view of the Data Base 
Administrator, and as such, it is the refer- 
ence manual for the whole system. It is not, 
though, intended to be a tutorial guide for 
beginning Data Base Administrators or 
DBMS-10 users. In addition, it is assumed 
that the reader has a knowledge of the 
COBOL language." 


(Comment) Two stumbling points are the 
word beginning and the phrase in addition. 
Beginning, the minor fault, should have the 
in front of it to prevent people from reading 
it as meaning starting something. In addition 
poses another problem. Since the previous 
fact was negative the reader is set for 
another negative fact when he reads "/« 
addition" And when he learns it isn't 
negative, he has to go back to verify that the 
mistake isn't his. 

(Example) "Occasionally, you may want to 
calculate a function, for example, the square 
of a number. Instead of writing a small 
program to calculate this function, BASIC 
provides functions as part of the language, 
some of which are described in Chapter 1." 

(Comment) Because the word writing modi- 
fies BASIC, the reader thinks BASIC can do 
two things: (1) write small programs to 
calculate and (2) provide a function. This he - 
realizes is illogical, so he has to back up and 
read again. 

2. The writer is addicted to abstract words, 
passive constructions, long words and vague 

(Example) "Sophisticated application speci- 
fications thus might require complex data 
structures involving many interrelation- 

(Comment) Sophisticated application speci- 
fications is long, abstract and vague, as are 
the words complex and interrelationships. 

(Example) "Many computer languages are 
currently in use, but BASIC is one of the 
simplest of these because of the small 
number of clearly understandable and easily 
learned commands that are required, its easy 
application in solving problems, and its 
practicality in an evolving educational 

(Comment) This sentence contains a handful 
of long, abstract terms that obscure its 


precise meaning: clearly understandable, 
readily learned, easy application in solving 
problems, practicality, and, last but by no 
means least, evolving educational environ- 
ment. In addition, the sentence is confusing 
at the word but. At that point the reader is 
set for a contrast involving the idea of being 
in use. He expects something like "but when 
BASIC goes to the field most of them will 
no longer be in use." 

3. The writing reveals an inner confusion of 

(Example) "Strings can be concatenated by 
means of the plus sign operator (+). The plus 
sign can be used to concatenate string 
formulas wherever a string formula is legal, 
with the exception that information cannot 
be stored by means of LET or CHANGE 
statements in concatenated string variables." 

(Comment) The exception talks about the 
storage of string variables; whereas the main 
assertion talks about the use of the plus sign 
to concatenate string variables. The excep- 
tion is irrelevant. 

(Example) "BASIC is a problem-solving lan- 
guage that is easy to learn and conver- 
sational, and has wide application in the 
scientific, business, and educational 

(Comment) The language says that these 
ideas ("easy to learn and conversational" 
and "wide application in the scientific, 
business, and educational communities") are 
equal. Actually, in thought they should be 
related as cause and effect. The sentence 
should begin with because or one of its 

A minor defect is seen in the words easy to 
learn and conversational. The phrase is 
choppy and lacking in rhythm. It serves to 
disconcert the reader momentarily. How- 
ever, if the writer puts the word conver- 
sational first, he can make that phrase flow 


for the reader: conversational and easy to 

(Example) "This software provides features 
which allow more than one run-unit to 
concurrently retrieve the data in the data 
base even while one run-unit is updating it." 

(Comment) Although this is a much less 
serious offense than the previous two, it still 
causes the reader to wonder what is going on 
in the writer's mind. Concurrently and even 
while have the same meaning. Writers should 
delete concurrently and even. 

4. The writer fails to put his main idea at the 
beginning of the paragraph. "Suspense," 
according to Thomas Johnson, "defeats the 
purpose of exposition. The information the 
reader needs to know should be at the 
beginning ..." 

(Example) "This mode should be used with 
caution. When this option is assembled in 
the module, GTRCAL and DISSKP are used 
in combination to determine what is to be 
done to the display file. This mode is 
provided for the careful user who has one or 
more curves being displayed and wishes to 
change one or more of them without turning 
off the display while it is being done. To do 
this, the user must first create the display 
file with GTRCAL set to 1 and DISSKP set 
to 1 in all INIT/DISPLAY tables. To recalcu- 
late the display file, set GTRCAL to -1 and 
for any INIT-DISPLAY table which is not to 
be calculated, clear DISSKP. DISSKP will be 
reset to 1 upon return. The user in effect is 
saying that the space required for recalcula- 
tion of each curve will not exceed that 
required initially . . ." 

(Comment) The first two sentences do not 
contain the main idea of the paragraph. The 
third one does. It should be the first 
sentence in the paragraph. 

The writer failed to simplify the opening 
paragraphs of his exposition. "In any piece 



of technical writing," Thomas Johnson says, 
"no matter how difficult, the opening para- 
graphs should be intelligible to every inter- 
ested reader. Don't avoid technical 
terminology, but simplify the language 
around it." 

(Example) (This is the paragraph labeled 
"Introduction" in Chapter 3 of a DEC 

"The AFC1 1 and AD01 devices are used for 
industrial and laboratory analog data 
acquisition. The AFC 11 is a flexible, high 
performance, multi-channel analog to digital 
(A/D) converter. Under program control the 
AFC1 1 performs a 13-bit A/D conversion at 
a rate of 200 channels/second. The AFC 11 
can multiplex a maximum of 1024 differen- 
tial input analog signals. The AD01 is also a 
multi-channel A/D converter; however, it 
differs from the AFC11, in that it multi- 
plexes up to 64 analog signals. In the 
following sections, the AFC 11 device 
handler task is discussed first, and then the 
AD01 device handler is described." 

(Comment) Actually, this paragraph of 
introduction causes more reader pain than 
the technical description of the AFC 11 
device handler task that follows it. 

6. The writer's presentation of detail is so 
dense that reader cannot absorb all the ideas 
at his normal reading pace. 

(Example) The paragraph in No. 5 illustrates 
this fault. 

(Comment) The word dense refers to the 
clustering of ideas. In the preceding para- 
graph density appears in such expressions as 
industrial and laboratory data acquisition, 
flexible, high-performance, multi-channel 
analog to digital (A/D) converter, and multi- 
plex a maximum of 1024 differential input 
analog signals. 



That's the substance of what the experts have to say 
about the writer. Although they may appear to condemn 
technical writing, that is actually not the case. They 
realize it has been performing an essential function for a 
long time. What they are saying is that it has obvious 
faults. They stress the need for improvement. Unani- 
mously, they urge you to sharpen your skill on the 
competence level; they suggest that you make technical 
writing more readable. 





"Who is my reader?" 

This question will plague you every time you begin a 
writing task. And rightly so. As we have already 
indicated, the reader is by far the most important 
element in a verbal communication system. Indeed, he is 
the reason it exists. He influences the language of the 
technical document, its format, illustrations, typography 
— everything about it. To ignore his needs, then, is to 
endanger your very prospects as a technical writer. On 
the other hand, if you create an accurate image of him, 
and choose your words well according to that image, 
you greatly increase your chances for success. 

In this chapter you will observe the reader as if he is 
under a microscope. You will see the two profiles of him 
created from the findings of many experts in the fields 
of reading and writing. As far as we know, this is the 
first time such information has been assembled in a book 
for writers. Most of it is scattered among some 20 
research projects conducted under the auspices of the 
U.S. Office of Education. 


George Klare, longtime reading researcher and author of 
strongly urges the writer to construct a detailed profile 
of the reader. Here are questions that Klare says can help 
you form this detailed reader image. 

1. Previous experience. What is the reader's 
previous experience with the subject? Be 
specific. Mark it down in years and depth of 
knowledge. Indicate depth of knowledge by 
showing exactly what he can do: Say that he 
can program in COBOL, FORTRAN, and 
SNOBOL rather than say that he has been a 


programmer for 4 years. Is it educational or 
work experience? Or a combination? Spell it 
out in kind and years. Does he understand 
technical terms or must you simplify them 
for him? Remember that material difficult 
for the general reader may be perfectly 
understandable by a person with a spe- 
cialized background. 

2. Educational level What was the last grade 
completed in school? This is crucial because 
a high correlation exists between grade 
completed and the individual's reading level. 
Thus the language chosen for a high school 
graduate would ordinarily be simpler than 
that chosen for a reader with a college 
education. Besides, readability tests give 
ratings in terms of grade level. So if your 
writing is to undergo a readability test, you 
should have the grade level of the reader 
firmly in mind before you write. If your 
book is targeted for the general reader, 
Klare's statistics on the median number of 
years of schooling completed by American 
adults could help you: 

1940- between 8 and 9 
1950- between 9 and 10 
1957- between 10 and 11 
1960- about 11 
1970 -around 12 

At any rate, he suggests that sometimes you 
may want to use U.S. Census data to get a 
better picture of the educational level of 
your readership. 

3. Motivation. Does he have a strong desire to 
read what you write? Does he have to read it 
to do his work? If so, he's likely to have a 
strong motive for learning. Or is he a 
voluntary reader likely to lay the book aside 
if the subject or style either bores him or 
forces him to expend too much effort in 
extracting the meaning. If you're trying to 
attract the voluntary reader, then you must 
use simpler language. He is not looking for a 
challenge and won't stay around long if your 
book offers him one. 


4. Intellectual level. Is the reader smart? How 
smart? Smarter than you? Smarter than the 
communicator in the next office? Take a 
guess at his I.Q. 110? 120? How do you 
know he's smart? What does he do that 
indicates high intelligence? Reads the clas- 
sics? Writes poetry? Plays chess? What does 
he do in his spare time? Granted, you can't 
answer all these questions. But by asking 
them, you can get a better notion of your 
reader's brain power. Don't forget that your 
words have to percolate through his mind in 
order to be understood. 

5. Interests. All the experts agree that you 
communicate more effectively when you use 
terms and examples familiar to the reader. If 
he's interested in inventories and payrolls 
and you give him examples from the world 
of finance and banking, you'll very likely 
lose him. Ask yourself: When he learns the 
subject, how is he going to apply it? This is 
one indication of his interests. 

6. Attention to the material. Where and under 
what circumstances will he be reading your 
document? If he is to read it in the 
computer room, seeking information to use 
in a panic situation, then lists, examples, 
outlines, boldface typography, and quick- 
reference tables will dominate in your 
presentation. However, you can resort to a 
different strategy if he can give it undivided 
attention in a relatively calm atmosphere. In 
this case you can furnish longer explanations 
and repeat information to aid his compre- 
hension. Exactly what is the reader's situa- 
tion? Visualize it. Write it down. Take it into 

Another expert, Herbert Michaelson, adds two more 
categories to the reader profile. 

7. Superior or subordinate to you in knowl- 
edge. You treat your subject matter differ- 
ently for different readers. The result of this 
attitude toward the subject matter is called 
the tone of your writing. Does the reader 


know less than you about the general field 
your topic is in? If he does, simplify it. 
However, if he's an expert — superior to you 
in knowledge — give him the full technical 

8. Administrator or technical person. Again 
tone is a factor. The administrator is inter- 
ested in concepts, significance, responsi- 
bilities, and potential. He is, in a word, your 
subordinate in technical knowledge. For him 
you must select and combine data to teach 
him what he needs to know to understand 
the software. The technical person, however, 
gets the full load of technical details — all he 
needs to know in order to use the software. 

And Tyler Hicks, prolific engineer-turned-writer, adds a 
directive to complete your reader image. 

9. Select a particular reader. From all the 
flesh-and-blood people you know, choose 
one person who represents the readership 
you're aiming at. Keep that person's name 
constantly in mind as you write. Paste it on 
the wall near your desk. Use him as the test 
of everything you write. Ask: Can he under- 
stand this? If not, rewrite it so he can. (And 
it wouldn't be a bad idea to let your typical 
reader review the entire manuscript when it's 

There you have all you need to create a good reader 
image. To be sure, you won't use all these categories for 
the template of your particular reader. You should use 
what you need. That's the spirit in which they are given 
— as principles to guide your imagination, rather than as 
rigid rules to hamper it. 


You can further sharpen your reader image with facts 
about readers in general: about how poorly they read 
and how inefficiently they process information. To be 
sure, it's a universal truth that all readers read differ- 
ently. But some of the facts supporting that truth show 
readers to be much less proficient than even the most 
pessimistic views have them. 


For instance, in 1970 J.R. Bormuth of the Department 
of Education at the University of Chicago wrote a 
monograph entitled "Illiteracy in the Suburbs." In it he 

1. That 51% of the students graduating from 
American middle-class suburban high schools 
cannot read and understand the textbooks in 
their classes, and 

2. That 41% cannot read and understand ordi- 
nary newspapers and magazines. 

Before looking at the implications of these statements, 
we want to emphasize two facts. Mr. Bormuth is no 
sensation-seeking newspaperman. He is a highly re- 
spected educator, a painstaking researcher, and a re- 
nowned interpreter of facts. Secondly, he has limited his 
observations to "middle-class suburban high schools," 
where the quality of education has usually been con- 
sidered better than in the teeming cities. In other words, 
he's talking about the so-called better schools. If he had 
included the big cities, the percentages would have been 

What do Bormuth' s percentages mean in concrete 
numbers? Well, each year American high schools grad- 
uate about 3 million students. (In fact, in 1969, the year 
immediately preceding Bormuth's report, our high 
schools graduated 2,839,000 students.) Of the nearly 3 
million graduates — remember that Bormuth's percent- 
ages would likely be higher if he considered the cities — 
at least half cannot read their textbooks. And what's 
even more jolting, these students are in a situation where 
they can get help in reading their texts! 

What does this mean for you as a technical communi- 
cator? At the outset, let's agree that not all these people 
went to work for factories and diners or joined 
construction crews. There's nothing. to indicate they're 
unintelligent. They just cannot read well. Doubtless 
many of them went on to become secretaries, clerks, 
technicians, computer operators, key-punch operators, 
programmer aides, etc. Aged 22 and 23 today, they 
probably form part of the readership of your manuals. 
Surely, the way events fall out, some of them will 
become managers and assistant managers of data centers. 


And when you add the number of illiterate graduates 
this year to that of last year and the year before, and so 
on, you should gather that at least half the adult 
population of the United States has trouble reading. 
Such people have a difficult time getting meaning from 
written material. They are likely to misunderstand big 
words and misinterpret long complex sentences. 

Another view of American readers, almost as startling as 
Bormuth's, was given by Helen R. Lowe in an article in 
the November 1959 issue of ATLANTIC MONTHLY. 
She worked for many years with superior tenth-, 
eleventh-, and twelfth-grade students. She found them 
making such disturbing reading errors as the following: 

Appearing in Book 

Word Read by Student 







grocery man 












God knows 

good news 













To darken the picture still more, she said this is just a 
sample from a list of some 100,000 errors by excellent 
students all the way up to the college level. Add to these 
misreadings the errors in comprehension arising from 
distorting the tenses of verbs and applying modifiers to 
the wrong words, and the picture of reader inefficiency 
gets even worse. 

Once again, what is the significance of all this to you as a 
communicator? Two observations are relevant. One is 
reinforcement of the notion that you cannot generalize 
about readers on the basis of your own good reading 
ability. Take it as gospel that readers are generally not as 
good as you are. In fact, they are much worse. 


The second point concerns your choice of words. 
Choose many-syllable words only if you want to run the 
risk of missing your reader. Granted, you as a writer 
have a perfect right to use words like inert, antiseptic > 
and hydride. But if your reader is likely to read them as 
inherent, adhesive, and hydroxide, what have you 
gained? One thing is certain, you've failed to inform the 
reader, you've failed to communicate. Anything you 
think you've gained is trivial beside the failure to 
communicate. A purely practical point shedding addi- 
tional light on this topic comes from the experiments of 
E.B. Coleman and C.R. Miller. They found that college 
sophomores learned the most from instructional ma- 
terials written on the fifth grade level. Moral: to 
communicate the most, use small words and short 

One thing we know with absolute certainty about 
readers is that they all get a somewhat different meaning 
from a piece of writing. This is so partly because each 
one judges your words by his own unique experience. 
Moreover, they adapt your words to their own expecta- 
tions, limitations, and ignorance. To illustrate this truth, 
Helen Lowe tells the story of the student who con- 
stantly read elephant house for elephant cage. The writer 
made a mistake, she said when asked for an explanation: 
elephants live in houses not cages. Another one con- 
tinually read supper for dinner, because people eat 
supper at night, not dinner. 

It would indeed be helpful if you could tell the good 
reader from the bad by looking at him. But this of 
course is impossible. You cannot even tell one from the 
other by observing their eye movements and reading 

As educator George Klare points out, in cases where 
experimenters used a camera, they could not differ- 
entiate the reading habits of normal people and the 
feebleminded. Both groups read at the same speed and 
with the same kind of eye movements. Only a compre- 
hension test can tell which is which. 


Another point to remember is that the grammar of the 
poor reader is not always the same as your grammar. 


Thus, your writing may be grammatically perfect. Yet it 
may fail to communicate because the inefficient reader 
habitually distorts the grammar. Donald Dansereau 
found this to be so in his research for the Department of 
Health, Education, and Welfare. (Helen Lowe said pretty 
much the same thing in her article in ATLANTIC 

Dansereau's readers - who go all the way' up to the 
college level — were insensitive to structural cues in 
written material. For example, they got no sense of time 
whatsoever from the -ed on a verb in a sentence like the 


He hurried to keep all his appointments. 

Perhaps the sentence would communicate better if the 
writer inserted the word yesterday. 

The net result is clearly pictured in the words of Helen 

"In addition to these spectacular distortions, most 
students make many less startling errors through 
constant deviations in tenses and pronouns and by 
omissions, interpolations, paraphrases, conjectures, 
and complete improvisations, so that paragraph 
after paragraph reaches their minds garbled, 
blurred, altered, vitiated — and ungrammatical." 


Researchers in reading and writing have proved in 
experiment after experiment that certain factors tend to 
interfere with a reader's understanding of written ma- 
terial. Some of the latest findings on this subject appear 
below. Others appear in Chapters 4 through 9 of this 

Long Sentences and Long Words 

Rudolf Flesch, Robert Gunning, George Klare, and 
others too numerous to specify are unanimous in 
identifying long sentences and long words as barriers to 
reader understanding. Gunning (in his TECHNIQUE OF 
CLEAR WRITING) implies that they waste the reader's 
energy so that he cannot grapple with the ideas involved. 
Klare makes the case that long sentences and long words 
slow down reading and tax the reader's memory. The 


length of the word affects word recognition, he says; so 
the longer the word the slower the word recognition 
time. This means the reader must invest more time and 
energy in learning what you are trying to say. Klare also 
contends that long sentences place a greater burden on 
the reader's memory span. The longer the sentence, the 
more the reader has to remember till he gets to the end 
of it. 

But the most telling indictment of long sentences and 
long words comes from a scholar of the 19th century: 
Herbert Spencer. In his PHILOSOPHY OF STYLE 
appears the following passage: 

"A reader or listener has at each moment but a 
limited amount of mental power available. To 
recognize and interpret the symbols presented to 
him requires part of this power; to arrange and 
combine the images suggested requires a further 
part; and only that part which remains can be used 
for realizing the thought conveyed. Hence, the 
more time and attention it takes to achieve and 
understand each sentence, the less time and 
attention can be given to the contained idea; and 
less vividly will that idea be conceived. . ." 

Sentence Structure 

Your sentence structure may serve to burden the poor 
reader. We will say more about this in Chapter 6, but 
here we want to relate some interesting observations by 
Ernst von Glasersfeld. He points out that a lot of 
split-second internal processing takes place when the 
reader sees a sentence that begins with the words: 

The coding she did. . . 

At this point the reader has to pull two possible 
interpretations into temporary mental storage. One of 
them is that the structure is a relative clause equal to 
"The coding that she did" in case the sentence is: 

The coding she did was praised by her supervisor. 

The other interpretation is that the structure is inverted, 
in case the sentence is: 


The coding she did but the documentation she 
refused to do. \ 

Von Glasersfeld labels such sentences "prospective am- 
biguity." They make for greater difficulty than normal 
sentences, he says, because two or more different 
interpretations have to be made and kept in temporary 
storage during processing of the sentence. Since they do 
indeed cause greater reading difficulty, the writer should 
question whether the reader can derive enough benefit 
to warrant their use. 

Of course, the sentence structure in the example appears 
infrequently in technical writing. But the same cannot 
be said of the conditional clause without the word if. It 
appears often and has just as much "prospective am- 
biguity" as von Glasersfeld's example. Here is one such 
sentence from a DEC manual: 

"Should the Data Base Administrator, or any user, 
suspect that a privacy key has become public 
knowledge, the Data Base Administrator should 
immediately issue a new unique lock/key com- 
bination to replace the one in question." 

At the word should the reader must be prepared for (1) 
a question or (2) a conditional clause. And he must keep 
these alternatives in mind until he reaches the word that. 
Truly, the conditional without if — especially if the 
clause is long — serves to burden the poor reader. 

Position of Information in a Paragraph 

Does the position of information in a paragraph influ- 
ence a reader's ability to remember it? The answer is an 
emphatic yes. This was the finding of Bonnie Meyer and 
George McConkie, who presented a paper on this topic 
in April 1974 at the annual meeting of the American 
Education Research Association. The experiments of 
Meyer and McConkie showed that information at the 
beginning of a paragraph is retained longer than that at 
the end. In fact, the deeper in the paragraph the 
information appears, the less likely it is to be re- 
membered. An additional finding is that topic sentences 
at the beginning of paragraphs are remembered longer 
than the supporting details following them. (Apparently, 
the details lose their independent identity in the reader's 


memory over a period of time and merge with the 
information in the topic sentence.) 

This information is important when you write tutorial, 
or instructional, material. Remember you defeat your 
own purpose when you put the most important data in 
the middle or at the end of a paragraph. 

Position of Adverbial Gauses in a Sentence 

An adverbial clause at the beginning of a sentence is 
more difficult to understand and remember than one at 
the end. Many researchers (notably H. Clark, K. Smith, 
L. McMahon, and V.M. Holmes) have proved, for 
example, that the first of these two sentences poses the 
greater comprehension difficulty to the reader: 

1 . When Tom arrived everybody cheered. 

2. Everybody cheered when Tom arrived. 

The reason for this really doesn't matter. (Suffice it to 
say it concerns the reader's memory span and the way 
the human mind processes information.) 

The important thing is its significance to you as a 
communicator: You reduce your chances of communi- 
cating when you place adverbial clauses at the beginning 
of sentences. 

Bearing in mind the information in the preceding 
section, you triply reduce your chances of communi- 
cating if: 

1 . You put adverbial clauses at the beginning of 

2. You make those adverbial clauses long. 

3. You fill them with long words. 

As Lois van Rooy points out, readability decreases as 
clause length increases. 

One Clause Instead of Two 

Experiments by E.B. Coleman in 1965 showed that the 
reader can understand information in two clauses better 
than he can in one. Thus, the subjects in his experiments 
understood sentence 1 much better than sentence 2: 


1. If you knew the Mississippi, it would be 

2. A knowledge of the Mississippi would be 

And to hark back to the statements in the preceding 
section, you can improve comprehension still more by 
revising sentence 1 as follows: 

1. It would be helpful if you knew the 

Once again, we want to emphasize that you compound 
the difficulty of sentence 2 when you add words. Thus, 
the writer of this sentence inflicts much pain on the 

2. A comprehensive knowledge of the salient 
facts about the Mississippi from its discovery 
to the extensive government dam-building 
projects of the 20th century would be 

Changes in the Logical Order of Your Exposition 

Experiments with college students by E.K. Darnell in 
1963 proved that changes in the logical order of a piece 
of writing interferes with comprehension. This means 
that you as a writer can expect a diminishing of reader 
comprehension if you alter the logical order you have 
led him to expect. For example, suppose someone had 
written a piece about the American Revolution for the 
Bicentennial. In writing about the events leading up to 
the outbreak of hostilities, he treated them in the 
following order: 

1. Stamp Act (1765) 

2. Declaratory Act (1766) 

3. Boston Massacre (1770) 

4. Boston Tea Party (1 773) 

5. Destruction of the Gaspee (1772) 

6. Intolerable Acts (1774) 

As soon as he mentions 1, 2, 3, and 4, with their dates, 
he has led the reader to believe he is going to follow 
sequential, or chronological, order. The reader believes, 
in other words, that the writer has made a contract with 


him to follow that order. The writer's move to 1773, 
then 1772, and then to 1774 disturbs his comprehen- 
sion. Later, the reader will experience difficulty in 
recalling facts 4, 5, and 6. 

Subject Complements Instead of Object Complements 
Look at these two sentences: 

1 . Your suggestion that the programmer should 
do the user manual seemed to alarm him. 

2. He seemed alarmed by your suggestion that 
the programmer should do the user manual. 

In Sentence 1 the underlined clause is called a subject 
complement. In Sentence 2 it is called an object 

Although the clauses are identical, the one in Sentence 1 
is, in the words of V.M. Holmes, "significantly more 
difficult" to understand. His explanation focuses on the 
occurrence of the main verbs. In Sentence 2, the main 
verb comes early, so you know the main idea of the 
sentence as soon as you reach the third word. In 
Sentence 1, the main verb comes late, so the reader has 
to carry from 10 to 12 words in temporary memory 
before the meaning of the sentence is resolved for him. 

Processing the main clause first, Holmes concludes, 
decreases perceptual complexity and increases 


Here are some tips on how you can help the reader 
comprehend your exposition. 

1 . Be logical in your presentation. Decide on a 
systematic order — and stick to it. 

2. Use explaining links such as because, there- 
fore, so, if . . . then, as a result, etc. They 
call attention to the logic of your thinking 
by pointing out a cause, means or result. 
(Chapter 4 tells you more about links, or 

3. Introduce examples frequently, especially 
where the material is difficult. 


4. Let daylight into your copy. Long para- 
graphs have an unfavorable psychological 
effect on the reader. Break them up. Maybe 
3 or 4 paragraphs a page would be a realistic 
goal. As Robert Gunning says: "Readers like 
periods and white space." 

5. Make your writing readable: write shorter 
sentences and smaller words. 

6. Use structural paragraphs throughout your 
document. Structural paragraphs serve to 
introduce, summarize, or review a topic for a 
reader. Usually appearing at the beginning or 
end of a section, they help orient the reader. 
They give perspective. 


Stated simply, the principle of least effort says that an 
individual will invariably choose to get to his goal by the 
route involving the least work. Applying this to reading, 
George Klare says the reader prefers readable material 
over the difficult because it requires less effort on his 
part. Klare observed that people will read two grades 
below their level when reading for pleasure. Sometimes 
they will tackle material above their grade level, but 
their motivation must be powerful. Usually this happens 
when the subject matter is very interesting or very 
necessary, as in the case of health information, or when 
the reader has a strong urge to learn. 

There's a moral here. Don't get the notion that you can 
"challenge" the reader with difficult passages. If he sees 
there's too much effort involved, he'll drop your book 
and get the information elsewhere. Perhaps he'll tele- 
phone the software specialist to get it orally. At any 
rate, you can be sure he'll take the easy road. 


That's the picture of the reader that emerges from the 
latest research. We hope we've made one fact about him 
stand out: generally he reads poorly. And the larger the 
readership you aim to reach, the greater the number of 
poor readers you'll meet. They certainly won't admit 
they're poor readers. Perhaps they don't even realize 
they are. So unless you take pains to help them in the 
ways suggested in this chapter, you'll miss them with 
your message. Then they'll have a perfect right to scream 


that your manual is garbage, because the burden of 
communicating rests squarely on you. 

When you stop to think about it, your job is much 
tougher than you thought. Not only must you write to 
be understood by the reader; but considering his many 
reading disabilities, you must also make certain you are 
not misunderstood. 





One sure way to help your reader is through strong 
organization of the paragraphs you write. This advice is 
especially apt when your ideas are totally unfamiliar to 
him or when his background in the subject is weak. In 
either case, the tighter you make the organization, the 
better the transfer of ideas. 

This view has the unqualified support of prominent 
professionals in the field of verbal communication. Two 
such professionals — Ludwig Mosberg and Fred Shima - 
tell of an experiment that typifies the beneficial effect 
of tight structure on communication. Although the 
experiment had to do with oral rather than written 
discourse, the findings are applicable to both. In the 
experiment, college students listened to speeches rated 
low, moderate, high, and very high in structure or 
organization. (Incidentally, the very high speeches con- 
tained transitional words and phrases and summaries of 
the main points.) After each speaker delivered his 
speech, the students took a 30-item test. Final results 
showed that all students, without exception, scored 
higher as the speech structure increased. 

This chapter tells you how to gain such organization in 
your technical writing. The chief emphasis is on two 
guiding principles: unity and coherence. In reading 
about these, you will learn about topic sentences, 
connectives, and the concept of motion. You'll also 
learn how to recognize the foremost example of unor- 
ganized writing: the haystack paragraph. All the 
examples used to highlight the discussion of unity, 
coherence, and the haystack paragraph are from DEC 
software manuals. 


The first sentence of every paragraph is a kind of 
contract between you and the reader. That sentence, in 


effect, tells him: "The paragraph will discuss the topic, 
or idea, presented in this sentence." When he reads the 
first sentence, he decides what the topic of your 
paragraph is. Thereafter, he expects you to live up to the 
contract: to stick to your topic. You break the contract 
— and risk losing his attention — when your first 
sentence discusses one topic, and any part of the 
paragraph discusses another. 

Thus, you focus his attention with the first sentence. 
Then you hold his attention by making your paragraph 
stick to the topic in that sentence. We call the first 
sentence the topic sentence. In writing other than 
exposition the topic sentence can appear anywhere in 
the paragraph. 

Consistency in sticking to the topic throughout the 
paragraph is called unity. Briefly, a paragraph is unified 
when all its sentences discuss the topic appearing in your 
opening sentence. When you give such unity to a 
paragraph you take the first step in organizing it. 

Here are some examples of good, unified paragraphs 
from DEC software documentation. 

"The programmer must write a program, which is 
a list of instructions for the computer to follow to 
arrive at a solution for a given problem. This list of 
instructions is based on a computational method, 
sometimes called an algorithm, to solve the prob- 
lem. The list of instructions is placed in the 
computer memory to activate the applicable cir- 
cuitry so that the computer can process the 
problem. This chapter describes the procedure to 
be followed when writing a program to be used on 
the PDP-8 family of computers." 

The first sentence tells us the paragraph will discuss a 
program, or list of instructions. This is the essence of the 
writer-reader contract in this paragraph. And the writer 
fulfills the contract not only by discussing the topic in 
each sentence but also by specifically mentioning it in 
each sentence. 


Another good example from the same manual: 

"A program written to perform a specific opera- 
. tion often includes sets of instructions which 
perform intermediate tasks. These intermediate 
tasks may be finding a square root, or typing a 
character on a keyboard. Such operations are 
often performed many times in the running of one 
program and may be coded as subroutines. To 
eliminate the need of writing the complete set of 
instructions each time the operation must be 
performed, the JMS (jump to subroutine) instruc- 
tion is used. The JMS instruction stores a pointer 
address in the first location of the subroutine and 
transfers control to the second location of the 
subroutine. After the subroutine is executed, the 
pointer address identifies the next instruction to 
be executed. Thus, the programmer has at his 
disposal a simple means of exiting from the normal 
flow of his program to perform an intermediate 
task and a means of return to the correct location 
upon completion of the task. (This return is 
accomplished by using indirect addressing, which 
is discussed later in this chapter.) The following 
example illustrates the action of the JMS in- 

Here the writer promises to talk about sets of instruc- 
tions which perform "intermediate tasks." And the 
promise is kept all the way through the paragraph. A 
new paragraph could have been started at "To eliminate 
the need of writing " but that would be a stylistic choice 
of the individual doing the writing. Actually, the 
material on the JMS instruction is wholly relevant 
because that instruction is intimately related to inter- 
mediate tasks. Notice how the writer repeated the term 
intermediate task in the third-last sentence to reinforce 
the unity of the subject matter. (The term was first used 
in Sentence 2.) Notice too that the one sentence that is 
slightly off the topic is properly placed within paren- 
theses. Notice, finally, how the last sentence carried the 
reader smoothly into the relevant example. 

It's too bad Robert Gunning did not come upon writing 
of this caliber before he wrote his TECHNIQUE OF 
CLEAR WRITING. It could have altered his undeniably 


pessimistic view of technical writing. And remember that 
we are now analyzing the paragraph for unity only. 
We're saying nothing about it being clear and exact. Nor 
do we mention the skillful use of transitional words 
(these, such, after, thus, etc.) to tie the whole paragraph 
together. Taking all these things into account, both 
Gunning and Thomas Johnson would have to conclude 
that this is good technical writing. 

The final example is a unified paragraph from a different 
DEC manual - this time a programmer's reference 

"PIREX (peripheral executive), a component of 
the UNICHANNEL-15 (UC15) Software System, 
is described in Chapters 3 and 4 of this manual. 
PIREX is a multiprogramming peripheral processor 
executive executed by the PDP-1 1 . It is designed 
to accept any number of requests from programs 
on the PDP-15 or PDP-11 and process them on a 
priority basis while processing other tasks concur- 
rently (e.g., spooling other I/O requests). PIREX 
services all input/output requests from the PDP-15 
in parallel on a controlled priority basis. Requests 
to busy routines (tasks) are automatically entered 
(queued) onto a waiting list and processed when- 
ever the task in reference is free. In a background 
environment, PIREX is also capable of supporting 
up to four priority-driven software tasks initiated 
by the PDP- 15 or the PDP- 1 1 ." 

The one defect in this paragraph is slight: for a moment 
the reader is in doubt as to the topic of the paragraph. It 
could be a thumbnail sketch of PIREX, or a short 
overview of Chapters 3 and 4. The first word in Sentence 
2, however, resolves the doubt. Again, as in the previous 
paragraphs, the writer makes a contract and abides by it. 

But sometimes we break faith with the reader. Look at 
the following paragraph, for example: 

"A computer, like any other machine, is used 
because it does certain tasks better and more 
efficiently than humans. Specifically, it can receive 
more information and process it faster than a 
human. Often, this speed means that weeks or 
months of pencil and paper work can be replaced 


by a method requiring only minutes of computer 
time. Therefore, computers are used when the 
time saved by using a computer offsets its cost. 
Further, because of its capacity to handle large 
volumes of data in a very short time, a computer 
may be the only means of resolving problems 
where time is of the essence. Because of the 
advantages of high speed and high capacity, 
computers are being used more and more in 
business, industry, and research. Most computer 
applications can be classified as either business 
uses, which usually rely upon the computer's 
capacity to store and quickly retrieve large 
amounts of information, or scientific uses, which 
require accuracy and speed in performing mathe- 
matical calculations. Both of these are performed 
on general purpose computers. Some examples of 
computer applications are given below." 

Analysis of this paragraph shows that the writer prom- 
ised in the topic sentence to talk about how a computer 
"does certain tasks better and more efficiently than 
humans." That's his contract with the reader, and in the 
second and third sentences he lives up to his obligation. 
Sentence 4, however, breaks the contract. It talks about 
computer use in relation to cost. The next sentence 
moves to still another topic: computer use in business, 
industry, and research. A wholly new topic — computer 
applications — confronts us in the last three sentences. 
These needless shifts of topic destroy the unity of the 
paragraph and disturb the reader. (They also interfere 
with reader comprehension. As a test of this last 
statement, reread the paragraph, then wait five minutes 
before trying to recall the facts in it.) 

Another instance of breach of contract appears in the 
following paragraph. 

"The Data Base Administrator must be aware of 
his organization's long-range plans as well as of 
long-range needs of the users. For instance, several 
groups of users might be formulating plans for 
data bases using interrelated data. The Data Base 
Administrator should be able to provide common 
access for these users. Understanding short-range 
user plans and needs as they pertain to specific 
application requirements is also necessary. A user, 


for instance, might require data from an existing 
data base which could possibly cause conflicts 
with the current users* interests. These differences 
must be reconciled. The Data Base Administrator 
has the responsibility to see that the data base is 
an effective, efficient tool for all users." 

Looking at the first sentence, you discover the topic to 
be the long-range plans of the administrator's organiza- 
tion and the long-range needs of the user. But Sentence 
4 introduces a new topic: short-range user plans or 
needs. What the writer needs to do to improve this 
paragraph is to move his last sentence up front. Certainly 
it wears the trappings of a topic sentence. Then he 
should change the second sentence to read: 

To do this he must be aware of both the 
long-range plans of his organization and the 
long-range needs of the users. 

The result will at least be a unified paragraph. 

The final paragraph comes from a different manual. 

"The text Editor (EDIT) is used to create and 
modify ASCII source files so that these files can be 
used as input to other system programs such as the 
assembler or BASIC. Controlled by user com- 
mands from the keyboard, EDIT reads ASCII files 
from a storage device, makes specified changes and 
writes ASCII files to a storage device or lists them 
on the line printer or console terminal. EDIT 
allows efficient use of VT-1 1 display hardware, if 
this is part of the system configuration." 

The breach of contract appears in the last sentence in 
this paragraph. There, the writer introduces "the effi- 
cient use of VT-11 display hardware" as a topic to 
compete with the one in the first sentence. 


A less elementary way of organizing paragraphs to help 
the reader is by means of coherence, or continuity. 
Coherence, from one point of view, refers to the smooth 
connection between the sentences in a paragraph. It is 
the flow or movement from sentence to sentence, which 
the reader senses as an ease in passing from one thing to 


another. To the reader the sentences are tightly linked, 
rather than loosely associated. He is therefore likely to 
describe such writing as "easy to read'* or "easy to 

Looked at from another point of view, coherence refers 
to the smooth flow of thought in written material. Seen 
as such, it is merely a verbal reflection of the linkage 
among the ideas in the writer's mind. Coherence in this 
sense is like thought itself. It has the lively onward 
motion of thought, as it carries the reader through the 
material. Thus, coherence is vitally connected with the 
competence level in writing. You can be effective in 
getting your ideas into the reader's head only when you 
show him how they are connected. 

And how does the writer show such connections? In 
other words, what are the transitions (i.e., words and 
expressions) you can use to guide the reader? 

The simplest transitions, of course, are pronouns and the 
repetition of keywords from preceding sentences. These 
are basic connectives that tie facts together but con- 
tribute little to the flow of ideas. They have value 
because by indicating identity or sameness of ideas they 
keep the reader oriented. But they are of secondary 
value in that they fail to show the reader the kind and 
order of your ideas. 

For the latter aim, you need the words and expressions 
listed in Table 1. 

As a quick test, why don't you go through a couple of 
pages of your own writing and underline any words 
appearing in Table 1? If you don't find them often 
enough, then you are not tying your ideas together 
logically and naturally. And your sentences, in the words 
of Jacques Barzun and Henry Graff are "weak at the 

Similarly, let's test the joints of some examples from 
DEC software manuals. 

"When speaking of memory locations, it is very 
important that a clear distinction is made between 
the address of a location and the contents of that 
location. A memory reference instruction refers to 


a location by a 12-bit address; however, the 
instruction causes the computer to take some 
specified action with the content of the location. 
Thus, although the address of a specific location in 
memory remains the same, the content of the 
location is subject to change. In summary, a 
memory reference instruction uses a 12-bit address 
value to refer to a memory location and it operates 
on the 12-bit binary number stored in the refer- 
enced memory location." 

There is no joint trouble in this excerpt. In the four 
sentences the writer guides the reader with seven of the 
words and expressions in Table 1. 

He uses when to denote time; however and although to 
show contrast between ideas; thus to indicate a cause- 
and-effect relationship; and, in two instances, to show 
equality of ideas; and in summary to signal the reader 
that a summary of the paragraph is coming his way. All 
these things considered, the paragraph is a good example 
of coherence in writing. 

Another model of effective coherence in DEC software 
writing is as follows: 

"Writing the above program was greatly simplified 
because mnemonic codes were used for the octal 
instructions. However, writing down the absolute 
address of each instruction is clearly an incon- 
venience. // the programmer later adds or deletes 
instructions, thus altering the location assignments 
of his program, he has to rewrite those instructions 
whose operands refer to the altered assignments. If 
the programmer wishes to move the program to a 
different section of memory, he must rewrite the 
program. Since such changes must be made often, 
especially in large programs, a better means of 
assigning locations is needed. The assembler assigns 
this better means." 

Here, in the skillful use of because, however, if, thus and 
since, the writer guides us through the ideas in his 
paragraph. The paragraph is coherent because it is the 
verbal reflection of the writer's ordered thinking. 


Table 1 
Some Transitional Words and Expressions 


Conjunctions and Adverbs 

























Adverbs of Time 










Simple Expressions 

another way to 

in essence 

instead of 

as well 

in fact 

more specifically 

at the same 

in general 

no matter what 


in other words 

on the other 

for the same 

in short 



in the same 

to begin with 

in addition to 


that's what 

in brief 

in sum 


in contrast 

in summary 

what's more 

Subordinate Conjunctions 










But not all the paragraphs in technical writing are as 
unified and coherent - that is, as organized - as the two 
you've just read. Indeed, many technical paragraphs are 
almost devoid of guiding words. And because they lack 
the organization that coherence supplies, they place the 


entire burden of interpretation on the reader. In effect, 
the writer says to the reader: "Here are the facts. You 
connect them." 

The typical example of such unorganized writing is the 
haystack paragraph. (It is also called the catalog or 
grocery list paragraph.) 

The Haystack Paragraph 

At DEC a short time ago a programmer returned a 
manual she had been reviewing to the supervisor of the 
writing group where it had been written. The supervisor 
asked her how she liked it. She was hesitant in her reply. 
"I don't know,** she said. "There's something wrong, but 
I can't put my finger on it." He asked her whether it was 
accurate. She said: "Oh, technically, it's all right. All the 
facts are there. But it's just. . . ." She paused and 
shrugged. "Well, you have to work too hard to read it." 

As soon as the supervisor looked at the manual he 
spotted the trouble: it was filled with haystack para- 
graphs. These are clusters of technical facts swept into 
neat paragraph piles by the technical writer who gives 
the reader little or no guidance. 

Here is an example of the kind of paragraph that appears 
often in DEC software documentation. 

"EXPAND is an RT-11 system program which 
processes the macro references in a macro assem- 
bly language source file. EXPAND accepts a subset 
of the complete macro language and, using the 
system library file SYSMAC.8K, produces an 
output file in which all legal macro references are 
expanded into macro-free source code. EXPAND 
is normally used with ASEMBL, the RT-1 1 as- 
sembler designed for minimum memory 

Here you have a catalog or list of facts with no attempt 
on the part of the writer to connect one idea to another. 
You can shuffle and rearrange these three sentences and 
they would be just as informative as they are now. What 
is the common relevance of these facts about EXPAND? 
If the writer had to supply a heading for this paragraph, 
would it be "Important Facts About EXPAND"? Or 
"Introductory Information About EXPAND"? Or. . .? 


As the paragraph now stands, the reader has to do all the 
interpreting - all the guessing. He has no way of 
knowing whether the list is partial or complete. Too, he 
doesn't know whether one fact is more important than 
the others. In a haystack paragraph all facts are 
grammatically equal. Thus, there is no emphasis, no 
selective impact on the reader. 

From a different software manual comes the following 

"Lab Applications-1 1 modules are available to 
perform operator console interaction, data acqui- 
sition, data editing, Fast Fourier transformation, 
output printing and displaying. Lab Applica- 
tions-1 1 allows many variations of these functions. 
The library of modules will be enhanced as time 
goes on, and as application needs are defined, 
more and more of the requirements for laboratory 
computing will be supplied by DIGITAL." 

Only one connective is used in the whole paragraph. 
That's the word these in the second sentence; and it 
serves to identify ideas, not to indicate any logical 
relationship among them. Replace these with the, and 
you can shuffle and rearrange these sentences and still 
retain the same effect — or lack of it. Again, there is no 
indication of relative importance, or of what all these 
facts mean to the individual reader, or even why they are 
assembled in this paragraph at this time. And as in the 
previous paragraph, the reader must assume the whole 
burden of interpreting the data — a task no reader of 
exposition should ever have to undertake. 

The next example is more frustrating to the reader 
because the logical relationships are implied, but not 
explicitly shown. 

**As these data bases increase both in use and 
sophistication, their creation and management 
must be relinquished by individual application 
programmers in favor of greater coordination and 
centralized control. It has become too costly 
and/or impractical, in most instances, for indi- 
, vidual programmers to create data bases for single 
applications on a one-to-one basis. The need at 
present is for data bases which are suitable for 


processing by, and available to, multiple ap- 

Three cause-and-effect relationships reside in this para- 
graph, but the writer uses no signals like because, since, 
thus, or therefore to reveal them to the reader. The 
reader is forced to dig them out — if he's interested. And 
if he's not interested, he misses the whole content of the 

Pause for a moment to consider how the writer of this 
paragraph defrauded the reader. At the time of writing 
he must have had ideas similar to the following, in his 

1. Because data bases are increasing in use and 
complexity, management by individual ap- 
plication programmers must give way to 
centralized control. 

2. Because the creation of data bases for single 
applications costs so much, individual pro- 
grammers will soon stop creating them. 

3. Because of the cost and the increase in use 
and complexity, programmers must create 
data bases that multiple applications can use. 

But when he wrote about the contents of his mind, he 
withheld the most important elements — the logical 
connections. Thus he made a mystery out of what 
should have been a revelation. The upshot is that he 
either overworks the reader or loses him entirely. 

Although the final haystack example is long, it is worth 
including because it contains most of the deficiencies of 
haystack writing. 

"Each line of the program begins with a line 
number of 1 to 5 digits that serves to identify the 
line as a statement. A program is made up of 
statements. Line numbers serve to specify the 
order in which these statements are to be per- 
formed. Before the program is run, BASIC sorts 
out and edits the program, putting the statements 
into the orders specified by their line numbers; 
thus, the program statements can be typed in any 
order, as long as each statement is prefixed with a 
line number indicating its proper sequence in the 


order of execution. Each statement starts after its 
line number with an English word which denotes 
the type of statement. Unlike statements, com- 
mands are not preceded by line numbers and are 
executed immediately after they are typed in. 
(Refer to Chapter 9 for a further description of 
commands.) Spaces and tabs have no significance 
in BASIC programs or commands, except in 
messages which are printed out, as in line 65 
above. Thus, spaces and tabs may, but need not be 
used to modify a program and make it more 

The writer attempted feebly — with these and thus and 
unlike - to connect parts of the paragraph. But his 
connectives were too few and too weak to alter the net 
reader impression that the paragraph is unorganized. 
Clearly his main trouble is that he started out in the 
haystack style and couldn't shake it entirely later on. 
Notice, for instance that his first three sentences contain 
a catalog, or list, of unconnected facts. This fault later 
leads him to include a discussion of statements, com- 
mands, and spaces and tabs in the same paragraph 
without attempting to show how they are related. 

The two uses of thus are especially weak. In the first 
instance, thus and the words with it do not expressly tell 
the reader that if he puts line numbers on his statements, 
BASIC will put them in order for him. On the contrary, 
he has to read through the sentence and then go back to 
piece together this information. The second use is much 
less effective. How tabs and spaces modify a program is 
not at all clear. And how they make a program "more 
readable" is vaguer still. The writer could greatly 
increase the effectiveness of both sentences by revising 
them to substitute because for thus. 


In the ideal communication system the writer provides 
paragraphs that are unified and coherent; the reader 
easily follows the writer's presentation, understanding all 
the ideas and their relationships. Granted, you'll never 
see such a system in operation. But if you follow the 
suggestions in this chapter for organizing your para- 
graphs to help the reader, you'll take a giant step toward 
realizing your part in it. You'll become a more effective 





All words are abstract. There are absolutely no excep- 
tions to this truth. This is the teaching of Alfred 
Korzybski, author of SCIENCE AND SANITY, which 
was first published in October 1933. Korzybski, a 
philosopher and scientist, gave us many new insights into 
the nature of language and meaning. Two of his 
contributions are discussed in this chapter. The first 
deals with the relationship between language and reality. 
The second has been popularized by his disciples as the 
"ladder of abstraction." 


There is no direct connection between language and the 
world around us. The connection is indirect - through 
the minds of the people who use the language. Because 
you and I agree that a particular word shall refer to a 
certain thing in the external world, we can use that word 
to communicate about that thing. Our agreement, then, 
is the only connection between the word and the thing. 
And such an agreement is purely arbitrary: we could just 
as easily have agreed that another word would stand for 
that thing. 

In Figure 4 a dotted line stands for the indirect 
relationship between words and the external world. 

The word dog BZ \C The animal dog 

Figure 4. Indirect Connection Between Word and Thing 


In the diagram, the writer (A) sees an animal (C). He 
uses the word dog (B) to refer to this animal. This word 
is what users of the English language have agreed shall 
refer to that kind of animal. If (A) were among people 
who knew only French, the word dog would have no 
meaning. (A) would have to know the word chien in 
order to communicate. Similarly, among Spanish- 
speaking people, (A) would have to know the word 
perro. In fact, the very existence of different languages 
proves there is no direct connection between words and 

So the connection between any word and what it refers 
to lies in the minds and experiences of the writer and the 
reader. The writer depends on the reader to relate the 
word to its referent. But what is clear to one class of 
reader may be as meaningless to another as the word dog 
is to people who know only chien or perro. 


As soon as we use a word, say dog, we start the 
abstraction process. For there is no such thing as dog in 
the real world. Out there exist "Fido" and "Rex" and 
"Prince" and all the other individual dogs with their 
individual names and individual characteristics. What we 
have done with the word dog is make it refer to certain 
selected characteristics of all dogs. In doing so, we left 
out other, specific, characteristics of individual dogs. 
This is the essence of the process of abstraction: we 
leave out certain specific characteristics. 

Thus, we use dog to mean, in the words of the 
ENCYCLOPEDIA AMERICANA, "a domesticated car- 
nivorous mammal remarkable for its intelligence and its 
attachment to man." What have we left out? For one 
thing, we say nothing about overall size, shape of head, 
or the color or kind of fur. Nor do we mention the 
sound "dogs" make, their odor, or their sleeping and 
eating habits. Too, we leave wholly out of account any 
notion of the dog as a living organism, ingesting food, 
transforming it, and getting rid of it. Moreover, we fail 
to focus on the dog as science sees it: a mass of swirling 
electrons. Hence, dog, as a word, becomes a shorthand 
term for a small subset of the total number of specific 
characteristics of all the dogs in existence. In sum, dog is 
a general word. 


The ladder of abstraction is a graphic means of showing 
(1) how we get more abstract and general as we ascend 
the ladder, and (2) how we get more concrete and 
specific as we descend the ladder. Figure 5, for instance, 
illustrates this in the case of dog. 

Notice that when we go up the ladder, each level refers 
to more items and says less about any one individually. 






Figure 5. Going Up the Ladder is More Abstract 

Carol's Duke, for example, refers to one specific member 
of the entire dog population. Up one rung on the ladder, 
dog designates any one or all of the approximately 100 
million dogs in the world. Still higher on the ladder, 
beast, can be applied to all animals except man — not 
only dogs, but also cats, bears, beavers, elephants, etc. 
On a still higher level of generality is animal, which 
designates any living thing that is not a plant. Living 
thing, on the topmost rung, is the most abstract, or 
general, term of all: it covers all plants and animals. 

There is an important lesson here for you as a writer: the 
higher your words on the abstraction ladder, the more 
meanings your reader can read into them. Remember 
what we said in the preceding section ("Language and 
Reality"): the writer depends on the reader to relate the 
word to its referent. Thus, you can use living things to 
refer to dogs, if you wish. But you risk having some of 
your readers apply it to plants, others to bears and 
coyotes, still others to men or fish. 

On the other hand, the lower you go on the abstraction 
ladder, the more specific you get. Thus, going down 
betters your chance of communicating. 


Look at the example in Figure 6. This time let's take a 
different perspective. Let's say that as we move down 
the ladder we become less abstract. 







Figure 6. Going Down the Ladder is More Concrete 

Suppose the public relations group at New-Look Com- 
pany issued a press release containing the following . 

New-Look expects its newly acquired asset to 
bring accuracy to its 1 inventory and speed to its 
payroll operation. 

The reader doesn't get much information from that 
sentence. The word asset can mean anything at all of 
value. So if we have no more concrete reference, we 
could think it referred to a building, a machine, or a 
system or organization. Going down the ladder, here's 
what information the other terms convey: 

chattel: Here we have taken a small step 

in the direction of communi- 
cating. This term rules out real 
estate but covers all other tan- 
gible possessions. 

machine: This covers any complex device 

for doing work. 

computer: Now we have zeroed in on a 

particular type of machine. But 
it could be the product of any of 
a score or so manufacturers. 

DECsystem-10 More specific still. We know it's 

computer: a DEC computer. But we don't 


know what it consists of; that is, 
we don't know its memory size, 
the number and kind of periph- 
erals, the specific pieces of soft- 
ware, etc. 
New-Look We have arrived at the bottom of 

DECsystem-10 the verbal ladder. Here is some- 
Computer: thing we can actually see and 

touch. Peripherals, configura- 
tion, size, software - all make 
this designation less abstract — 
more concrete — more specific. 

In summary, the higher you climb the ladder of 
abstraction, the more general you get. The lower you go, 
the more specific you get. Stay low on the ladder for 
more efficient communication. For when you're low, 
there's less chance of the reader's projecting erroneous 
meanings into your words. 

Abstract Words vs. Concrete Words 
In everyday language we use the terms concrete and 
abstract to refer to the different kinds of words we 
speak or write. Generally speaking, an abstract word 
refers to something that does not exist in the physical 
world; that is, it does not have a specific referent against 
which it can be checked. Thus, component and system 
and implementation are abstract words in this sense. So 
are facility, maintenance, establishment, honesty, evil, 
beauty, and configuration. Notice that a number of 
these words refer to qualities, ideas, conditions, or 

Concrete words, on the other hand, do have physical 
referents. And as we have seen in the discussion of the 
ladder of abstraction, concrete words are either specific 
or general. They are specific when they refer to 
particular things in the physical world: Carol's Duke, 
Senator Kennedy, the Declaration of Independence, 
Digital Equipment Corporation. They are general when 
they refer to a class of persons or things: dog, hill, 
programmer, physicist, computer, newspaper. 

In regard to communication, the consensus of the 
experts is that concrete words are by far the more 
effective. As Robert Gunning says: "Words that stand 


for things we know by the senses are safest in communi- 
cation because there is greater possibility that you and 
the reader will have a large area of overlap in experience 
with these words." And Lois Van Rooy, in her advice to 
writers of books, echoes Gunning: "Write in terms as 
concrete and specific as possible." 

In summary, use concrete words for better communica- 
tion. As an aid in finding concrete words use the 
following three tests: 

1. Can this be seen or touched? 

2. Does this exist in the physical world? 

3. Is this picturable? (A picturable term names 
something that the reader can visualize as 
having size and form. The word car is 
picturable. The word experience is not.) 


Words high on the ladder of abstraction are useful as 
tools of thought. Since you can pack them with any 
number of details of your own choosing, you can use 
them as a kind of logical shorthand. For example, you 
can make the word sophistication stand for 5 years of 
programming experience with MACRO, a working 
knowledge of FORTRAN and ALGOL, and two years of 
supervising experience in data base management. Clearly, 
it is much more economical for you to use the word 
sophistication than the specific details it stands for. The 
same can be said for any abstract word, say, optimiza- 
tion or enhancement or quality. As long as the term has 
definable meaning for you as the thinker, you can use it 
with benefit in your thinking. As an additional bonus, 
you can very likely organize your thoughts better 
because you are dealing with fewer terms. 

But abstract words are usually no good for communica- 
tion. So the trick is to use them for thinking and then 
translate them into more concrete terms when you 

If, however, you must use an abstract word in your 
writing, use it as a shorthand way of referring to details 
you've already given. For example, suppose you have 
already described for your reader a DECsystem-10 
computer consisting of the following: 150K of memory, 
a line printer, 2 tape drives, a disk, 3 terminals, a 


monitor, and PIP, DDT, MACRO and TECO. Therefore, 
you can use the abstract word system instead of 
repeating the elements listed in the preceding sentence. 
Actually, you have made system concrete in this 
instance by giving it precise meaning. It now refers to 
something that can be seen in the physical world. And 
this is the only way of communicating successfully with 
abstract words. 


Writing that contains a high percentage of abstract nouns 
reveals the following faults: 

1 . It is difficult to understand. 

2. It is inefficient in transferring information 
from the writer's head to the reader's. 

3. It is difficult to memorize. 

These findings were reported by E.B. Coleman and C.R. 
1968. In sum, these educators aim the following message 
at the technical writer: you cannot communicate suc- 
cessfully by using a lot of abstract nouns. 

The reason for this is clear. Words high on the ladder of 
abstraction are vague in meaning. They tend to make a 
passage general and indefinite. Unaccompanied by con- 
crete nouns and active verbs, they cloak a passage in a 
smog-like haze of ambiguity that hides the writer's 
intended message. 

Since no word — even the most specific of concrete 
terms — means the same thing to all readers, the best we 
can say about an abstract style is that it means different 
things to different readers. Whereas the worst we can say 
is that it means nothing at all to any readers. 

Take the Word freedom for example. Obviously, it 
means one thing to a reader of COSMOPOLITAN and 
another to a reader of TRUE. And those meanings 
would perhaps fail to square with the meaning given it 
by the reader of GAY WORLD MAGAZINE or the 
MAGAZINE. Similarly, a writer who uses the abstract 
word vehicle to convey the notion of Conestoga wagon 
. runs the risk that his reader thinks of a buckboard when 
he reads it, or even a phaeton. 


"Abstractions," in the words of Thomas Johnson, "are 
detrimental to exposition, because they are open to 
interpretation." Thus, they defeat the purpose of infor- 
mative writing, which should say the same thing to all 
the readers it's aimed at. Jokingly, Johnson suggested 
that when you don't know what you're trying to say, 
throw in a few abstract words. Further, if you want to 
bury an idea like a needle in a haystack, slip it into some 
abstract words. 

Morris Freedman in his SEVEN SINS OF TECHNICAL 
WRITING puts abstractions under the sin labeled "gen- 
eral fuzziness in communication." He, too, urges the use 
of specific concrete terms that say much the same thing 
to all the target readers. 

Two big dangers in the use of abstractions were pointed 
out by Robert Gunning and Herbert Michaelson. The 
former said that the writer gets so used to abstract words 
that after a while such words seem concrete to him. 
Thus, the writer is unaware of any need to question 
whether his words are fuzzy or misleading to the reader. 
What's even worse, he's writing prose only he under- 
stands. Michaelson pointed to a similar danger on the 
part of the reader. Most of the time, he said, the reader 
does not see the words as ambiguous. He just interprets 
them according to his own understanding and experi- 
ence, and then goes on his way. This kind of faulty 
communication - where the minds of the reader and the 
writer meet only through the wildest of coincidences - 
can easily result in accidents with equipment, loss of 
sales and of good will and, quite possibly, lawsuits. 

Our own documentation has many examples of the use 
of abstract terms whose meaning the reader is allowed to 
supply as he wishes. At least, the writer seems to have 
failed to give a meaning for him. In this vacuum the 
reader's mind will provide a meaning from his own 
experience. Here are a few of those examples from DEC 
software documentation. 

1 . (Example) "VERIFY provides the user with 
a facility for verifying the consistency and 
validity of a file structure on a specified 


(Comment) Facility, consistency, and valid- 
ity are high on the ladder of abstraction. 
Notice that they can be neither sensed nor 
pictured. Nor are they readily seen in the 
physical world. Such words are likely to 
leave the writer and reader poles apart 
regarding their meaning. 

2. (Example) "Therefore, while the scientist 
designs experiments or tests from content- 
area principles of his discipline, DIGITAL 
has designed implementation conventions, 
processes, and functions from computer 
principles. The processes and functions are 
the software elements of Lab Applications- 
11. This manual, as well as the whole nature 
of Lab Applications-11, is a set of system 
conventions or framework used in imple- 
menting applications. This generalized sys- 
tem design structure allows the user to 
create data handling systems which reflect 
experience the user personally may not have 
had time to acquire." 

(Comment) This entire paragraph is on a 
very high level of abstraction. How much of 
the writer's meaning is actually communi- 
cated can be known only by a quiz of both 
the writer and a typical reader. Certainly, 
ambiguity and emptiness appear in words 
and phrases on every line. For example, does 
content-area principles have an exact mean- 
ing? How about such abstractions as imple- 
mentation conventions, processes, computer 
principles, nature, system conventions, 
framework, applications, generalized system 
design structures, systems, and experience! 
Because of such abstract words and expres- 
sions, trying to pin down the meanings of 
words in this paragraph is like trying to put 
uniforms on a squad of ghosts. 

3. (Example) "Many computer languages are 
currently in use, but BASIC is one of the 
simplest of these because of the small 
number of clearly understandable and 
readily learned commands that are required, 


its easy application in solving problems, and 
its practicality in an evolving educational 

(Comment) Here, again, the meaning is 
elusive and constantly changing because of 
the abstractions the reader has to pour 
meaning into. Look, for example, at appli- 
cation, problems, and practicality. All three 
are vague and ambiguous. But for abstrac- 
tions that mean little to anybody, consider 
"evolving educational environment" and 
"clearly understandable and readily learned 
commands." Commands, of course, is a 
technical term but the qualifiers disturb its 
set technical meaning. Understandable to 
whom? What's the difference between 
understandable and learned in this phrase? 
Are the commands "readily learned" be- 
cause they are "clearly understandable?" 

4. (Example) "DOS/BATCH has been designed 
to manage resources efficiently and to be 
easy to use. It is suitable both for regular 
production work, and by reason of its data 
storage facilities and debugging aids, for 
program development. However, while it 
provides a response time fast enough for 
most requirements, it is not intended for use 
in on-line real-time data acquisition ap- 

(Comment) In this paragraph you'll find a 
number of abstract words people think have 
meaning because they've gotten used to 
them. High-level abstractions like resources, 
facilities, aids, requirements, and applica- 
tions still mean different things to different 
people. This paragraph is an excellent ex- 
ample of a case where the reader doesn't 
sense ambiguity. He merely supplies his own 
meaning and moves on. 

5. (Example) "Lab Applications-1 1 is an inte- 
gration of software, hardware, and theory 
which provides the laboratory computer user 
a combination of resources to solve his 
particular laboratory application problem." 


(Comment) Try, if you can, to state the 
meaning of "integration of hardware and 
theory." Then try "integration of software 
and theory." In this way you'll get some 
idea of the emptiness of the entire phrase 
"integration of software, hardware, and 

6. (Example) "Consider a research company 
which is organized into departments. Some 
of these departments are actively engaged in 
research; others are support departments, 
e.g., accounting, payroll, legal, etc. Each 
research department is broken up into re- 
search project groups, although there are 
some special projects which overlap depart- 
ments. The research projects are supported 
by contracts and grants which come from 
external sources — some projects receiving 
funds from more than one contract or 

(Comment) At first glance, this paragraph 
seems more concrete than the others, But it 
is still on a high level of generality. Witness 
words like departments, projects, research, 
research project groups, and grants. The 
writer failed to give these specific meaning, 
even though he intended this description as 
an example in his manual. And what can we 
say about the word funds? Surely the writer 
could have stepped down the abstraction 
ladder and used the word money. It's more 
exact — and more picturable. 


"What can I do to improve an abstract style?" you may 
ask. The experts (specifically Rudolf Flesch, Robert 
Gunning, Jacques Barzun, Henry Graff, Thomas John- 
son, E.B. Coleman and C.R. Miller) offer a number of 

1 . When you are writing on a very high level of 
abstraction, you can communicate more 
effectively, according to Flesch, by adding 


restatements on a more concrete level. This 
"corresponds to the use of illustrations, 
practical applications, examples, parables, 
and the like in teaching." In a word, Flesch 
advises you to back up abstract writing with 
concrete illustrations. 

(It is interesting to consider why the readers 
of technical manuals constantly clamor for 
more illustrations. Is it because the concrete 
examples make the abstract text more 

2. Gunning says you should always be aware of 
the position of your words on the ladder of 
abstraction. "Make certain that the terms 
you use actually have a connection at the 
ladder's bottom to observable facts." 

Remember the loose generality of the phrase 
"integration of software, hardware, and 
theory." If you are prone to write Jike that, 
ask yourself: What do these words refer to in 
the physical world? If the answer is nothing, 
then ask yourself: What am I trying to s'ay? 
What am I trying to tell the reader? 

3. Similar to Gunnings' s advice is Jacques 
Barzun's and Henry Graffs suggestion that 
you constantly ask: "Can this be touched or 

4. Coleman and Miller say that much writing is 
difficult to understand because the writer 
chooses verb-nominalizations instead of 
verbs. A verb-nominalization is an abstract 
noun derived from a verb. For example, 
operation, construction and creation are 
verb-nominalizations derived from the verbs 
operate, construct, and create, respectively. 
Coleman and Miller conclude that the writer 
can make his sentences more meaningful by 
reducing the number of verb-nominaliza- 
tions. You do this by changing the verb- 
nominalization into an active verb. Thus, 
you would change "Peter's creation of the 
program" to "Peter created the program." 

5. Gunning, a very articulate proponent of the 
use of short words, suggests that you use big 
words sparingly. Most big words, he points 
out, are abstract. They stand for qualities of 


things and relationships among things. Short 
words, on the other hand, are concrete: they 
refer to persons, places, things, and acts. 
Thomas Johnson asks you to spot fuzzy 
words in your own writing and then either 
eliminate them or give them specific mean- 
ing. Typical words that tend to make your 
style abstract appear in the following list: 





























































You will frequently have to use abstract terms in your 
technical writing. In doing so, you should realize that 
the farther you get from the concrete, the greater your 
chances of missing the reader. Therefore, your big need 
is always to know where you are on the ladder of 
abstraction so that you can intelligently question 
whether your terms will be understandable to the reader. 
When the level of abstraction is too high, consider using 
concrete examples to give meaning to your abstract 
words. Too, never stop trying to find a concrete 
replacement for the abstract term. As Rudolf Flesch 
pointed out, concreteness is even more important for 
successful communication than short sentences. And the 
reason is that there is an excellent chance you and your 
reader will have an overlap in experience of the things 
concrete words stand for. 





Traditional grammar defines a complex sentence as one 
that contains an independent clause and one or more 
dependent clauses. Thus, the following sentences are 
properly labeled complex: 

1 . He came when he was ready. 

2. He said that he would come when he was 

But this definition gives little indication of the real 
nature of a complex sentence. For one thing, it deals 
only with the surface appearance of the sentence. For 
another, it leaves the reader entirely out of account. 

The complexity discussed in this chapter is what the 
reader encounters when he tries to extract meaning from 
a piece of writing. In this sense, complexity refers to the 
total number of relations he must unravel to understand 
a passage. This definition, then, sees complexity as a 
composite of a number of factors. 

One of these factors is, of course, the surface structure 
of the sentence, for that is what the reader sees first. 
Actually, you as a writer can present the same content in 
a number of different surface structures. But the 
structures differ in effectiveness; that is, the reader can 
extract information more easily from one form than 
from another. For example, both of these sentences have 
the same content: 

The programmer drove his car to work. 

The car was driven to work by the programmer 
who owned it. 

The first sentence, however, is likely to prove more 
effective from the reader's viewpoint than the second. 


Another factor is the cluster of relationships beneath the 
surface. Ernst von Glasersfeld, a typical proponent of 
this view, analyzes this kind of complexity into the 
following parts: 

1. The occurrence of unfamiliar relationships. 

2. The prospective ambiguity of certain 

An illustration of unfamiliar relationships is seen in the 
following sentences about the Americans and the 

1. Americans fought with the Japanese in 
Peking in 1900. 

2. Americans fought with the Japanese in 
Manila in 1945. 

In the first sentence the Americans and Japanese are 
related as allies, whereas in the second they are related as 
enemies. And the difference is due to the relationship set 
up by the preposition with. In sentence 1 it means "on 
the side of — an unfamiliar usage; and in sentence 2 it 
means "against.'* Structurally, these sentences are identi- 
cal — each has an independent clause and three 
prepositional phrases. 

Von Glasersfeld's concept of prospective ambiguity was 
discussed in Chapter 3. There, we saw that the sentence 

The coding she did but the documentation she 
refused to do. 

required more reader effort to understand. For this 
reason von Glasersfeld called it complex. Traditional 
grammar, on the other hand, labels it compound — a 
designation that says nothing about its complexity. 

This, then, is the mark of complexity as discussed in this 
chapter: It makes the reader work hard to get meaning 
from a sentence. And when he must struggle too much 
with complexity, he cannot process the -information 
properly for recall later. 

What does all this mean to you as a technical writer? The 
answer is simple: Make your sentences less complex in 


order to communicate successfully. In this chapter you'll 
find a discussion of required memory level and sentence 
length in their relationship to complexity. And the 
closing paragraphs offer you some suggestions for the 
cure of this major ill of technical writing. 


One way to understand sentence complexity is to view it 
as a burden on the reader's memory. Jackson Morris 
coined the term required memory level to describe this 
burden. Required memory level, as he uses it, refers to 
the reader's capacity to remember facts and relationships 
as he reads. More specifically, it means the maximum 
number of facts he must remember at any point in the 
sentence. Thus, the larger the number of facts to be 
remembered, the greater the complexity of the sentence. 
In other words if the required memory level is high, the 
sentence is difficult for the reader. The more facts the 
reader must remember, the harder he must work to 
understand your writing. 

As a simple example, look at this sentence. 

The fifth house on the right on Dan's street is 
being sold. 

The reader has to remember four facts (fifth, house, 
right, and Dan's street) before he reaches the verb, which 
clarifies the subject for him. That is, when he gets to the 
verb, he knows what the subject means in this sentence. 
This is an important aspect of required memory level: it 
is usually equated with the distance between subject and 

Many word puzzles, Morris points out, are based on 
required memory level. The following one has been 
around a long time. 

"If a hen and a half lays an egg and a half in a day 
and a half, how many eggs does one hen lay in a 

Another familiar oldie asks the reader to identify the 
person who says these words while looking at a picture. 

"Brothers and sisters I have none. 

But this man's father is my father's son." 


Some technical writing imposes a burden as great as that 
of the word puzzles. Here is an example from one of our 

"Systems for digitizing and storing analog data, 
sorting and averaging multiple analog signals, and 
sequencing processing events, are all examples of 
laboratory problems where DIGITAL has knowl- 
edge independent of scientific discipline.'* 

In this case the reader must temporarily store and 
remember 1 3 facts before the predicate lets him know 
what they mean. Admittedly, there are individual 
variations in the capacity to cope with a required 
memory level. But retaining 13 facts would be a 
prodigious feat for even the most retentive of memories. 
Just for the fun of it, reread that sentence and, when 
you reach the word problems, try to remember the three 
problems. This exercise puts you in the place of the 
reader and tells whether you can cope with the required 
memory level. 

Another burdensome sentence from DEC software docu- 
mentation reads as follows: 

"By dividing the DBS files into pages, and storing 
selected records on these pages, and using a page as 
the basic I/O buffer size, DBCS operations that 
affect the records on only one page can be handled 
with a single disk access." 

This 42-word sentence requires the reader to grasp 19 
facts before he reaches the verb. And to make his burden 
even heavier, he's forced to retain 14 facts before he 
even gets to the simple subject (DBCS operations). 
Translated into other terms, this means he must read 24 
words before he gets his first inkling of what the 
sentence is about. You can test your memory, too, on 
this one, if you wish. But it really isn't necessary. 
Obviously, the required memory level is impossibly high. 
This is a first-rate example of how sentence complexity 
can help inhibit communication. 

Although the next example is shorter than the previous 
ones, it is by no means less complex. In fact, its 
complexity is so fascinating that on one occasion 12 


college graduates were given this sentence as a test. They 
had to read it, then wait five seconds before stating all 
the facts it contained. Not one of them was able to do 
this successfully. Here it is for your edification: 

"The establishment and maintenance of relation- 
ships between records specified by means of 
declaring sets in the schema, is a responsibility of 

To be sure, there is no known way to measure required 
memory level. But this shouldn't deter you from using it 
to your advantage as a communicator. Accept the fact 
that it does indeed exist. It is real because there is a limit 
to the temporary complexity that any human mind can 
adjust to. Hence you can help the reader by anticipating 
his difficulty with sentences having too great a distance 
between subject and verb. Revise such sentences before 
they get into the manual. Revert to the subject-verb- 
complement word order of the declarative sentence. This 
order not only reduces the required memory level, but it 
also clarifies relations among the various sentence 
elements. For example, say declaratively: 

DBMS-10 is responsible for setting and keeping up 
relationships between the records declared by sets 
in the schema. 

Transformational grammar presents a view similar in 
effect to Morris' required memory level. Coleman and 
Miller, who represent the transformational approach, 
speak of kernels as the simple units of language. Writers 
transform these kernels to make complex sentences. As 
an example of a single-kerneled sentence they offer the 

They gave her a dollar. 

On the other hand, a many-kerneled sentence is 

They admired her intelligent lectures. 

This sentence is made up of three kernels: she lectured, 
she was intelligent, and they admired. 


The more kernels a sentence contains, the more complex 
it is. And what is even more important for the 
communicator, Coleman and Miller's experiments 
proved that passages with few kernels (i.e., short 
sentences) are more easily understood by the reader. 
Again, because the few-kerneled sentences are less 
complex, the reader does not have to work so hard to 
unravel their meaning. 


Sentence length is a basic measure of complexity in a 
piece of writing. At least this is the view of George Klare 
and other longtime professionals in the study of reada- 
bility. And all the practical tests of readability so far 
devised - notably Flesch's Reading Ease Formula, 
Gunning's Fog Index, and the Dale-Chall Formula - 
embody sentence length as a factor. 

But it is not length per se that causes difficulty, for as 
we've just seen, a simply worded sentence of 22 words 
can engender as much reader pain as a rambling sentence 
of 42 words. Rather, it's the number of relationships in 
the sentence that causes the trouble. For this reason, we 
cannot say that a 30-word sentence is always twice as 
hard to understand as a 15-word sentence. We can say, 
however, that since relationships depend on words, long 
sentences are usually harder to understand simply 
because they contain more words. Further, we can indict 
long sentences because they strain the reader's memory 
span — his ability to recall material correctly after 
reading it once. 

Thus, the writing of informative sentences requires much 
skill on the part of the technical writer. In fact, it 
demands much more from you than most people realize. 
Not only must you give the reader the facts he needs, 
but you must also put it into chunks small enough for 
him to digest. Yours, then, is a job of constant 
adjustment: adjusting your technical material to sen- 
tence lengths the reader can understand. And you must 
do this more or less instinctively, on the basis of what 
you know about the reader and the complexity of the 
subject matter. 

How long should sentences be? The experts say that no 
one can put a ceiling on sentence length. They also say 


that sentences don't have to be of a uniform length. 
They advise you to adopt the standard of average 
sentence length. That is, some sentences will be very 
long, some short; but on the average they should be 

Rudolf Flesch in THE ART OF PLAIN TALK and 
Kenneth Houp and Thomas Pearsall in REPORTING 
TECHNICAL INFORMATION are quite specific in 
relating number of sentence words to ease of under- 
standing. Here is Flesch's correlation of number and 

Number of Words 

Reading Ease 

8 words or less 

Very Easy 




Fairly Easy 




Fairly Difficult 



29 words or more 

Very Difficult 

To summarize Flesch's findings, he says that an average 
length of 17 words makes for good readability. Houp 
and Pearsall, on the other hand, suggest a 21 -word 
average as a good target to aim for. 

Reducing sentence length doesn't mean you'll confine 
your presentation to simple sentences. No, you'll still 
balance and mix simple sentences with compound and 
complex, though the proportion of simple sentences will 
doubtless increase. What you will do, however, is reduce 
the number of subordinate clauses, phrases, and other 
modifiers in each sentence. 

To illustrate how our documentation would shape up 
against the Flesch and Houp-Pearsall measures, let's 
analyze some passages chosen at random. Here are the 
paragraphs; the analyses follow. 

1. "These procedures might include, but are 
not limited to, relationships between trans- 
actions and data bases, and responsibilities 
of the users and data base personnel. In 
order to fully accomplish all the tasks 


assigned to him and his staff, the Data Base 
Administrator should schedule time when 
the data base is not available to users in 
order to perform reorganization procedures 
(such as reassigning areas to different 
devices/media), FAILSAFE procedures, and 
other "house-cleaning" chores. This time is 
in addition to the normal preventive mainte- 
nance time for the computer system as a 
whole, and the users must be made aware of 
these activities. Likewise, users who merely 
want to retrieve information from the data 
base should know when this information is 
being updated by other users so that fresh 
copies of data are always available." 
(Average sentence length = 33 words) 

2. "Any format conversions and I/O control 
routines not needed in the resident section 
but required by overlay sections must be 
forcibly loaded into the resident section. 
This can be done by declaring the appro- 
priate globals in an assembly language 
routine or inserting dummy FORMAT and 
Input/Output statements in the resident 
main program for all those routines needed 
in the overlays and not required in the 
resident section. See Section 9.10 for further 
details." (Average Sentence Length = 26 

3. "The entry existence indicator is set nonzero 
when a buffer entry is made. When a 
requester has removed or processed an entry, 
it must clear its existence indicator in order 
to free the buffer entry position. Entries are 
made in a circular fashion, starting at the top 
(low address) filling in order of increasing 
memory addresses to the bottom (high 
address) and wrapping around from bottom 
to top. If input data occurs in a burst 
sufficient to overrun the buffer, data is 
discarded and a count of data overruns is 
incremented. The nonzero entry existence 
indicator also serves as an overrun indicator. 
A positive value (+1) indicates no overruns 
between entries, and a negative value is the 
two's complement of the number of times 
data has been discarded between entries. 


Word zero of the buffer is used by the 
handler task as a pointer into the buffer 
where the next set of interrupt information 
is to be entered. It is expected that the 
connected task maintains a pointer to that 
location in the buffer where it is to next 
retrieve contact interrupt data. When a task 
is triggered by the handler, it should process 
data in the buffer starting at the location 
indicated by its pointer and continuing in a 
circular fashion until the two pointers are 
equal. Equality of pointers means that the 
connected task has retrieved all the contact 
interrupt information that the handler has 
entered into the buffer. The pointer main- 
tained by the handler is to be thought of as a 
FORTRAN index into the buffer, i.e., the 
first location of the buffer is associated to 
the number (index) 1. The second location 
associated to the module number indicates a 
module on which a change of state in the 
direction of interest has been recognized for 
one or more discrete points." (Average 
sentence length = 25 words.) 

4. "The detection of any card reader error 
condition in Batch signifies a 'Device not 
Ready* state which elicits a A002 message 
and disables the reader interrupt. If the 
operator issues a CONTINUE command to 
resume processing, the error processor will 
recall the Transfer routine to repeat the read 
and exit to await the next interrupt. The 
operator is given the opportunity to correct 
the error before entering the CONTINUE 
command. The card using the error should 
be replaced and the replacement card should 
be the first card read when processing 
resumes. An exception to this procedure 
occurs whenever the A370 diagnostic 
message is printed. In this case, the last card 
from the output side should not be replaced 
in the input hopper for it has already been 
read." (Average Sentence Length = 21 

5. "Three indirect command files for system 
generation are provided in RSX-11D. Either 
32KGEN.CMD or 44KGEN.CMD can be 


used to define the system to Phase 1 instead 
of typing directives in response to Phase 1 
requests. SYSBLD.CMD is always used to 
perform Phase 2. If the user wishes to tailor 
the system to a particular installation, he can 
either edit the existing files or create new 
ones." (Average Sentence Length = 14 

Flesch's formula would give a high readability rating 
only to Paragraph 5 ; Houp and Pearsall would find only 
Paragraphs 4 and 5 acceptable. For a quick comparison 
of the paragraphs, see the data listed below. 


Sentence Houp-Pearsall 

Paragraph Length Flesch Rating Acceptance 

1 33 words Very Difficult No 

2 26 words Difficult No 

3 25 words Difficult No 

4 21 words Fairly Difficult Yes 

5 14 words Fairly Easy Yes 


Complex writing can hardly be called good informative 
writing because it makes the reader work too hard for 
his information. So anything you can do to reduce 
complexity is a step in the right direction. With that 
thought in mind, here are a few suggestions to help cure 
this major ill of sentence complexity. 

1. Use this as a guiding principle: No reader 
should ever have to analyze one of your 
sentences in order to understand it. 

2. When you have to write a long sentence, put 
the predicate close to the subject. 

3. Break compound sentences in two once in 
awhile. And then begin the second sentence 
with and, but, or for. 

4. See if you can use a period where you'd be 
inclined to use a comma or a which. 

5. Count the words in any long sentence. If 
there are more than 21, consider revising it. 


6. Use balanced, or parallel, phrasing to make 
long sentences more understandable. Con- 
sider that this 73-word sentence from 
Lincoln's Second Inaugural Address is under- 
standable because of its balance or 

"With malice toward none, with charity for 
all, with firmness in the right, as God gives 
us to see the right, let us strive on to finish 
the work we are in, to bind up the nation's 
wounds, to care for him who shall have 
borne the battle, and for his widow and 
orphan — to do all which may achieve and 
cherish a just and lasting peace among 
ourselves and with all nations." 

(The balance in that sentence is seen in the 
three phrases starting with the preposition 
with and the three infinitive phrases: to bind 
up . . . to care ... to do.) 

7. Imitate the natural structure of speech. One 
result is that your subjects and verbs will be 
close together. 

8. Constantly ask yourself this question when 
you revise: Can I write this sentence more 
briefly and clearly? 

9. Be encouraged by Robert Gunning's observa- 
tion: "Nearly any subject can be discussed in 
prose that does not go beyond mid-high- 
school level in complexity." 






Many experts on writing aim their heaviest verbal 
artillery at the use of the passive voice. Morris Freed- 
man, for example, calls it the "deadly passive" and labels 
it Sin Number 6 in his famous article "Seven Sins of 
Technical Writing." He asserts that it makes "any 
reading matter more difficult to understand, to get 
through, and to retain." In a similar vein, Herman 
Struck, writing in the SCIENTIFIC MONTHLY, advises: 
"... A writer might well consider every passive sick until 
he proves it healthy." 

This chapter also takes issue with the passive voice. But, 
as the title clearly indicates, "overuse," not use, is the ill 
that needs treatment. In other words, we do not 
condemn the passive voice per se. It can and should be 
used to good effect in technical writing — where, the 
emphasis is more often on the thing rather than on the 
person. What we do condemn, however, is the technical 
writer's needlessly heavy dependence on it. This, we 
submit, is what helps give technical writing the dead- 
liness that Freedman detects: a lack of emphasis, 
vitality, and motion. 

This chapter discusses (1) identification of the passive 
voice, (2) the case against the passive, (3) the case for 
the passive, (4) the passive in literature and in technical 
writing, and (5) suggestions on how to shift from the 
passive to the active. And as was the case in the 
preceding chapters, examples to illustrate the principles 
in the text are taken from DEC software documentation. 


In grammar the term voice refers to the way the subject 
is related to the action expressed by the verb. If the 
subject is the doer of the action, then the verb is in the 


active voice. On the other hand, if the subject is the 
receiver of the action, the verb is in the passive voice. 

A sentence, with its verb in the active voice usually has 
an object as well as a subject. The object is always the 
receiver of the action expressed by the verb. The verbs in 
the following sentences are in th e active vo ice. Notice 
that the subject is enclosed in a | rectangle | , the active 
verb is underlined once and the object twice . 

1- I John| hit the ball. 

2. The first five | instructions] initialize the 

3. Output | devices I record the results of com- 
puter operations. 

4. The arithmetic | unit | of a digital computer 
performs the actual work of computation 
and calculation. 

5* I You 1 verify the contents of memory with 
the EXAM switch. 

Here is a summary of the outstanding features of the 
active voice: 

1. The subject (John, instructions, devices, 
unit, you) performs the action expressed by 
the veib. 

2. The line of action in the sentence is direct. 
That is, the subject does the verb's action on 
the object; and movement goes forward 
from the beginning of the sentence to the 


Doer of Action — *-Verb — *- Receiver of Action 

3. The forward movement in the sentence is 
similar to that in thought or in actuality. 
(This is what gives the active voice its 
naturalness and force.) 

4. Since the subject spot is the emphatic spot, 
emphasis in the active sentence is on the 
doer of the action. 


The word passive conies from the Latin passivus, which 
means suffer. So in the passive sentence the subject is 
the sufferer, or the receiver, of the action. If you want 
to include the doer of the action (the person or thing 
that inflicts the suffering) you must put it after the verb 
in a phrase starting with the preposition by. Thus, the 
underlined verbs in the following sentences are in the 
passive voice: 

1 . John was jut by the ball. 

2. The loop is initialized by the first five 

3. The results of computer operations are 
recorded by output devices. 

4. The actual work of computation and calcula- 
tion is performed by the arithmetic unit of a 
digital computer. 

5. The contents of memory are verified by the 
EXAM switch. 

Here is a summary of the important points about the 
passive voice. 

1 . The subject of the verb in the passive voice is 
the receiver of the action. In the preceding 
sentences, for instance, the following are 
receivers of the action: John, loop, results, 
work, and contents. 

2. The passive verb always has two parts: 
(a) some form of the verb be (e.g., is, am, 
are, was, were, has been, have been, had 
been, etc.) and (b) the past participle of 
another verb. Usually the past participle 
ends in -ed, as is the case in sentences 2, 3, 4, 
and 5. Sometimes, however, it doesn't end in 
-ed, as in Sentence 1. 

3. The line of action in the passive sentence is 
backward, the reverse of what it is in 
thought or actuality. (This is why readers 
find passive sentences more difficult to 
understand than active sentences: they must 
search out the doer of the action.) 


Receiver of Action -* — Verb -* — Doer of Action 


4. Since the subject spot is the emphatic spot, 
emphasis in passive sentences is on the 
receiver of the action. 


The experts indict the passive voice on the following 
eight counts. 

Weakness — Robert Gunning and Thomas Johnson use 
such words as inert, motionless, inactive, and dull to 
describe verbs in the passive voice. And when you 
consider the role of the verb in the sentence, you can 
readily see their point. The verb is usually thought of as 
the power house of the sentence, the focus of action, 
energy, vitality, and movement. So widespread is this 
thinking that to speak of a passive verb sounds like a 
conflict of terms. In fact, most people, if given the word 
verb in a free association test, would doubtless respond 
with action words like run or push or climb or attack. 
Rarely would anyone come up with one of the pallid 
passives that abound in technical writing: 

is accomplished 
is achieved 
is allowed 
is arranged 
is assigned 
is attained 
is based 
is considered 
is defined 

is designated 
is established 
is effected 
is employed 
is enabled 
is facilitated 
is featured 
is involved 
is made 

is observed 
is performed 
is permitted 
is provided 
is required 
is specified 
is used 
is utilized 

So the first point in the case against the passive is that it 
usually weakens the force, or impact, of the sentence. 
And you can sense the general truth of this contention 
by observing the feebleness of the passive in these 

1 . No moss is gathered by a rolling stone. 

2. Early in the morning a utility pole was 
crashed into by a speeding car. 

3. Except where specified, each line typed by 
the user must be entered to the computer by 
typing the RETURN key. (DEC manual) 

4. The worm is caught by the early bird. 


5. The international dateline was crossed by 
the airplane. 

6. This was immediately realized by me. 

Complexity - A second point is that the passive is 
harder to understand than the active. The reader feels 
that the passive sentence is unnatural. Its subject is the 
receiver of the action, rather than the doer. And this, 
says researcher Alvin Granowsky, disconcerts the reader 
because he expects the subject to be the doer. Thus with 
the passive he has to keep track of the doer — a chore 
the active does not impose. To all this, Dan Slobin, 
AND VERBAL BEHAVIOR in 1966, adds: the negative 
passive is even more complex than the positive. Slobin 
tested subjects who ranged in age from 6 to 20. All of 
them found the negative passive the hardest of all 
sentences to understand. 

Concealment — Experts find the passive defective 
because it often fails to identify the doer. Some critics 
call this the case of the lost subject or the missing agent. 
It would be amusing if the failure did not sometimes 
have serious consequences: 

All doors in this building must be locked before 
you leave work for the day. 

Who's going to lock them? Chances are that no one will, 
because the directive does not specify the doer. At other 
times, the missing agent poses no serious problem. 

"Once these statements are mastered, the user can 
investigate the more advanced applications of 
these statements and the additional statements and 
features explained in Parts II and III." (DEC 

In this instance the reader can leap without strain to the 
conclusion that the user is the doer. But why not make 
the doer explicit and thus remove all uncertainty? 

After mastering these statements, the user can 
investigate the more advanced applications of 
these statements and the additional statements and 
features explained in Parts II and III. 


Misconstruction - The passive increases the likelihood 
of grammatical errors. This is especially so in the case of 
dangling modifiers. A modifier is said to dangle when, 
logically, it has nothing in the sentence to modify. For 
example, the underlined modifiers in the following 
sentences are dangling: 

1. In packing the car, his typewriter was 

(Typewriter is packing the car? Yet type- 
writer is the only word that packing can 
modify. To correct the sentence, change the 
verb to the active voice and put the logical 
doer in the subject spot: In packing the car, 
we overlooked his typewriter.) 

2. To examine the television set, the plug 
should be removed. 

(Plug is going to examine the television set? 
But plug is the only word that to examine 
can modify. Again, part of the correction of 
this fault is to put the logical doer into the 
subject spot: To examine the television set, 
you should remove the plug.) 

3. Flying bis helicopter close to the ground, the 
two-year-old wanderer was spotted in the 

(A two-year-old wanderer is flying? Once 
again we have a dangling modifier: Flying has 
nothing but wanderer to modify. And as 
before, we can correct the sentence by 

(1) putting the verb in the active voice and 

(2) inserting a logical doer in the subject 
spot: Flying his helicopter close to the 
ground, the pilot spotted the two-year-old 
wanderer in the woods.) 

In most cases the danglers aren't really misread. People 
smile at them and then quickly apply their own logic to 
extract the right meaning. However, when a passage 
contains a lot of them, the reader has to slow his pace 
needlessly. Consequently, he could become bored with 
the book and put it aside. And there is always the 
possibility that a dangling modifier may convey erro- 
neous — perhaps disastrous — information. 


Our documentation, with its emphasis on the use of the 
passive voice, contains many danglers. Here are a few 

1. "By dividing the DBS files into pages, and 
storing selected records on these pages, and 
using a page as the basic I/O buffer size, 
DBCS operations that affect the records on 
only one page can be handled with a single 
disk access/* 

(DBCS operations is dividing, storing, and 

2. "To clarify this, the concept of public disks 
and private disks must be explained." 

(The concept is clarifying this? If so, why 
must it be explained?) 

3. "In describing each of these statements, a 
line number is assumed and brackets are 
used to denote a general type." 

(Line numbers and brackets are describing 
the statements?) 

4. "To indicate that a statement is to be 
continued, the line is terminated with the 
LINE FEED key instead of the RETURN 

(The line is doing the indicating? Why not 
the user?) 

5. "This statement is used when writing a 
program to process data to be supplied while 
the program is running." 

(Who is writing a program? Not statement.) 

6. "Before considering data access, several as- 
pects of data organization should be 

(Aspects are considering?) 

7. "CTRL/C is typed by holding down the 
CTRL key while typing the C key." 

(We have two danglers and a verb - but no 
doer. When actions occur without an actor, 


the whole scene seems as eerie as the 
movement of the keys on a player piano.) 

Another grammatical confusion connected with the 
passive is the misreading caused by the pronoun it. One 
example from our documentation will illustrate the 

"The GOTO statement is used when it is desired to 
unconditionally transfer to some line other than 
the next sequential line in the program." 

(Not until the reader gets to the word to does he 
realize that it does not refer to GOTO statement. 
Frequent occurrences of the passive with it as the 
subject throws the reader off stride. This is 
especially so when the structure appears deep in 
the sentence, as it does in this example. For the 
more words that it can refer to, the greater the 
effort the reader must expend to extract the 

Wastefulness — Some critics fault the passive because it 
requires more words. Houp and Pearsall, for example, 
state flatly that the passive is uneconomical because it 
costs more to print more .words. 

Let's apply the principle of economy to some sentences 
in our own documentation. Each sentence is shown as it 
appears in the manual. The revision follows it. And 
finally, there appears a statement of the saving in words 
gained by the revision. 

1. (Original) "A READ statement is used to 
assign to the listed variables those values 
which are obtained from a DATA state- 
ment." (20 words) 

(Revision) A READ statement assigns to 
listed variables the values obtained from a 
DATA statement. ( 1 4 words) 

(Saving) From 20 words to 14: A saving of 6 

2. (Original) "The value assigned to a variable 
does not change until the next time a LET, 


INPUT or READ statement is encountered 
that contains a new value for that variable or 
when the variable is incremented by a FOR 
statement." (39 words) 

(Revision) The value assigned to a variable 
does not change until another LET, INPUT, 
or READ statement assigns a new value to 
that variable or until a FOR statement 
increments the variable. (31 words) 

(Saving) From 39 words to 31 : A saving of 8 

3. (Original) "TABS, like spaces, are used to 
make a program easy to read." (12 words) 

(Revision) TABS, like spaces, make a pro- 
gram easy to read. (9 words) 

(Saving) From 12 words to 9: A saving of 3 

4. (Original) "An expression is a group of 
symbols which can be evaluated by BASIC." 
(13 words) 

(Revision) An expression is a group of 
symbols BASIC can evaluate. (10 words) 

(Saving) From 13 words to 10: A saving of 3 

5. (Original) "The exclamation mark is nor- 
mally used to terminate the executable part 
of a line and begin the comment part of a 
line." (22 words) 

(Revision) The exclamation mark normally 
terminates the executable part of a line and 
begins the comment part. (16 words) 

(Saving) From 22 words to 16: A saving of 6 

6. (Original) "If the relationship is found false, 
then control is transferred to the statement 
following the IF statement (the next sequen- 
tially numbered line)." (22 words) 

(Revision) If the relationship is false, then 
control transfers to the statement following 


the IF statement (the next sequentially 
numbered line). (20 words) 

(Saving) From 22 to 20 words: A saving of 2 

7. (Original) "The COS-300 2780 Remote Data 
Communications Package (RDCP) is de- 
signed to provide remote users of IBM 360 
and 370 systems with both on-site proc- 
essing and remote job entry (RJE) com- 
patible with the IBM 2780 Data Trans- 
mission Terminal Model 1 ." (39 words) 

(Revision) The COS-300 2780 Remote Data 
Communications Package (RDCP) provides 
remote users of IBM 360 and 370 systems 
with both on-site processing and remote job 
entry (RJE) compatible with the IBM 2780 
Data Transmission Terminal Model 1. (36 

(Saving) From 39 words to 36: A saving of 3 

Wordiness — Both Freedman and Thomas Johnson say 
that the passive voice and awkward wordy constructions 
seem to go together. For example, the passive verb is 
assumed usually calls for on the part of. Too, the passive 
is questioned drags the lengthy as to whether along with 
it. Finally, as we have seen in the examples in the section 
on "Wastefulness," the passive is used and the infinitive 
invariably in its wake add up to four words that can 
easily be reduced to one. (Notice that the extra words 
the passive draws into the sentence are idle as far as 
meaning goes and serve only to obscure the words that 
do the work.) The net result of the passive and its wordy 
accompaniment is an unemphatic, verbose style. 

Some examples of that kind of style appear below. 

1 . (Original) "Debugging of the program can be 
achieved through the use o/DDT." 

(Revision) The user can debug his program 
with DDT. 


2. (Original) "The user is advised that the 
following hardware is required in order to 
run this software." 

(Revision) The user needs the following 
hardware to run this software. 

3. (Original) "A working knowledge of COBOL 
is assumed on the part of the programmer 
who wants to understand this manual." 

(Revision) The programmer must know 
COBOL to understand this manual. 

Deemphasis - The passive helps to bury the action of 
the real verb in the sentence. Not all passives do this, but 
the pallid passives listed under "Weakness" certainly 
have this as a built-in fault. Again, we turn to our 
documentation for sentences to highlight this fault. 

1 . (Original) "It can be used to synchronize the 
central processor to external events, count 
external events, or measure intervals of time 
between events. It can be used to start an 
analog to digital converter with the firing of 
a Schmitt trigger." 

(The real verbs in this passage are synchro- 
nize, count, measure, and start. The revision 
below eliminates the passive verbs and ele- 
vates the action words from their gram- 
atically inferior position as infinitives. 

(Revision) It can synchronize the central 
processor to external events, count external 
events, or measure intervals of time between 
events. It can also start an analog to digital 
converter with the firing of a Schmitt 

2. (Original) "It is commonly required to retain 

the final data in a readily accessible medium 

(The real verb in this sentence is retain. The 
revision promotes it from its infinitive status 
and thus puts the action in the verb spot 
where it belongs.) 


(Revision) The user should retain the final 
data in a readily accessible medium . . . 

Abstractness — Finally, the experts say that passive verbs 
attract abstract nouns. Thus, in the final analysis, the 
passive and the abstract work together to deprive the 
reader of the meaning he looks for in the passage. 

In the following examples, you can see how the passive 
allies itself with abstract nouns. 

1. (Original) "The practice of documenting 
software is disliked by many programmers.*' 

(Passive verb: is disliked. Abstract noun: 

(Revision) Many programmers don't like to 
document software. 

2. (Original) "The capability of the system to 
be used for solving laboratory problems is 
stressed in this manual." 

(Passive verb: is stressed. Abstract noun: 

(Revision) This manual stresses the use of 
the system for solving laboratory problems. 

3. (Original) "The chief measure of the soft- 
ware's success is indicated by the absence of 
SPR's in the last three months." 

(Passive verb: is indicated. Abstract noun: 

(Revision) The absence of SPR's in the last 
three months proves that the software is 


Despite what some experts seem to imply, the active 
voice is not always more effective. There arc many 
occasions when the passive is much the better choice. 

Of course the decision in regard to use rests with the 
writer. And it depends on what you want to emphasize. 


If you want to stress the receiver and take the emphasis 
off the actor, then you should turn to the passive. Too, 
if the doer of the action is unknown or unimportant, 
you should use the passive. The following sentences, for 
example, are more effective in the passive. 

1. The data center was wrecked by vandals last 

(Comment: Emphasis is on the receiver of 
the action.) 

2. Abraham Lincoln was defeated for the U.S. 
Senate by Stephen A. Douglas. 

(Comment: Emphasis is on the receiver. 
However, in a paragraph about Douglas, we 
would use the active voice and make 
Stephen A. Douglas the subject.) 

3. The twelve bits of the accumulator are 
numbered to 11. 

(Comment: Doer is unimportant.) 

4. The input and output units are combined in 
Figure 2-1 because in many cases the same 
device acts as both an input and an output 

(Comment: Emphasis is on the receiver; doer 
is unimportant.) 

5. Winter wheat is harvested in Saskatchewan. 
(Comment: Doer is unimportant.) 

6. The airplane was loaded in Saigon yesterday 

(Comment: Doer is unknown, very likely 

7. The first instruction of the subroutine is put 
in the second location of the subroutine. 

(Comment: Doer is unimportant.) 

8. Rubouts are not stored in the text buffer 
but are inserted by the Editor following all 
tab characters on the output tape. 

(Comment: Emphasis is on the receiver.) 


A validation error is flagged when the 
information written is not the same as the 
information read in. 

(Comment: Doer is unimportant.) 


Thomas Johnson in his book ANALYTICAL WRITING 
says that active verbs predominate in good literature. In 
Lincoln's "Gettysburg Address," he points out, only 3 
verbs out of the 30 are passive. In Hamlet's "To be, or 
not to be" there are only 2 passive verbs out of 25. And 
in the Twenty-third Psalm (King James Version) not one 
of the 1 5 verbs is passive. 

Quite in contrast to these figures are the proportions in 
technical writing. 

Three pages in DEC software manuals sampled at 
random show the following breakdown: 









(Sept 1974) 












To show how indiscriminate use of the passive results in 
unnatural prose, Johnson "passivized" the Twenty-third 
Psalm. To be sure few people would ever write some of 
the sentences he produced, but the passage can serve as 
an extreme to remember when you monitor your own 

The Lord is designated as my shepherd; nothing 
shall be wanted by me. 

I am made by him to lie down in green pastures; 
I am led by him beside the still waters. 


My soul is restored by him; I am led in the paths 
of righteousness for his name's sake. 

Yea, though the valley of the shadow of death is 
walked through by me, no evil will be feared by 
me; for I am accompanied by thee; I will be 
comforted by thy rod and thy staff. 

A table is prepared by thee in the presence of my 
enemies; my head is anointed with oil by thee; my 
cup is made to run over. 

Surely I will be followed by goodness and mercy 
all the days of my life; and the house of the Lord 
will be dwelled in by me forever. 


"Shifting to the active from the passive automatically 
improves a writer's style." This is the view of authors 
Menzel, Jones, and Boyd in their book WRITING A 
TECHNICAL PAPER. But don't interpret this quotation 
to mean improvement comes when all passive verbs are 
changed to the active. The passive, we repeat, has a place 
in technical writing, where the doer is often unimpor- 
tant. Thus, you should accept the statement by Menzel 
and his colleagues as merely a guide to improvement. 
Don't try to eliminate the passive; strive rather to curb 
its overuse. Improvement in communication comes when 
all needless passives are changed to the active. 

Here are some specific suggestions for shifting from the 
passive to the active: 

1 . When revising, ask of each sentence "Where's 
the action?" If the verb in the sentence is 
one of pallid passives listed in the section on 
"Weakness," you know the action isn't in 
the verb. Look for an active verb to substi- 
tute for the pallid passive. 

2. In trying to find an active substitute for the 
passive, look for an infinitive right after the 
passive verb. Change the infinitive to an 
active verb. 

This is what we did to the following sen- 
tence in the section on "Wastefulness." 


(Original) A READ statement is used to 
assign to the listed variables those values 
which are obtained from a DATA statement. 
(Infinitive is underlined.) 

(Revision) A READ statement assigns to the 
listed variables the values obtained from a 
DATA statement. (Active verb is under- 

3. Or you may locate the active verb in a noun 
made from a verb: 

(Original) When this option is assembled in 
the module, GTRCAL and DISSKP are used 
in combination to determine what is to be 
done to the display file. (Noun made from 
verb is underlined.) 

(Revision) When this option is assembled in 
the module, GTRCAL and DISSKP combine 
to determine what is to be done to the 
display file. (Active verb is underlined.) 

4. Watch the proportion of passive verbs in 
your writing. Robert Gunning says that the 
proper balance for effective communication 
is one passive to every 9 active verbs. This 
seems rather lopsided for technical writing — 
where 3 to 9 throughout a manual would 
appear more reasonable. 





The scene is the corner of Boylston and Tremont Streets 
in Boston. A new visitor to the city stands before the 
entrance to the MBTA subway. He wants to go to 
Copley Square to get a close-up view of the architecture 
of Trinity Church, the Boston Public Library, and the 
Old South Church. Above the entrance to the subway he 
sees a solitary sign: 


From his position, he can see another subway entrance 
across the street. This one is likewise labeled with a 
single sign: 


Which way to Copley Square? Is Copley "in" or "out"? 
The signs fail to enlighten him. To answer the question 
he must cither ask one of the Bostonians whizzing past 
him or descend the 50-odd steps into the subway to ask 
the person in the change booth. 

These two signs are examples of what is called fast 
pacing in writing - reliance on a few words to convey a 
lot of meaning. Fast-paced writing is terse, compressed, 
and compact. 

But no matter how it is described, it can always be 
identified by a sometimes disturbing characteristic: it 
leaves much information unsaid. Thus, in assuming a 
large background of knowledge on the part of the 
reader, such writing often leaves him more baffled than 
informed - with more questions than the text can 
answer. Look, for instance, at what the sign INBOUND 
TRAINS ONLY does not say: 


It does not tell the new visitor that this entrance 
leads to cars marked "Park Street'* and "Lech- 
mere"; or that "Park Street" cars go just one stop 
(to Park Street) and then he'll have to get off; or 
that "Lechmere" cars go to Park Street and 
beyond - to Government Center, Haymarket and 
Lechmere; or that he can alight at Government 
Center and make connections with rapid-transit 
cars to East Boston; or that he should take the 
entrance across the street to get to Copley Square. 

All this information is implied in that one sign. But there 
is no way this reader can know any of it — let alone all 
of it. Without someone to supply such details, he is lost, 
immobilized, unable for a time to proceed any further. 

Many times the fast pacing of technical writing does the 
same thing to the reader. For example, look at the three 
instructions given below. They are taken from a DEC 
software manual for relatively new users. 

1. Turn the Teletype control to LOCAL (see 
Figure C-l). 

2. Feed the blank tape into the punch. 

3. Depress the LOCK "ON" control. 

Step 2 is akin to INBOUND TRAINS ONLY. It doesn't 
tell the reader (l)how to start feeding the tape and 
(2) how to verify that he has fed it correctly. If someone 
should say that the reader of this manual should know 
these things, then the obvious question arises: Why is 
there an illustration to help him with steps 1 and 3 but 
nothing to help him with the seemingly more difficult 
second step? The existence of the figure indicating the 
controls shows that the writer considers the reader to be 
without experience. On the other hand, the terseness of 
Step 2 assumes that the reader does indeed have 
experience. The confusion of such fast pacing in 
technical writing could leave the reader as frustrated and 
immobile as the visitor in Boston reading INBOUND 
TRAINS ONLY for the first time. 

From another point of view, fast pacing refers to writing 
that packs a lot of technical data into few words. Seen 
on this level, fast pacing is called density. 


This chapter takes a close look at density in technical 
writing: at what it is, at the various forms it takes, at its 
effects, and at the ways it can be cured. 


Density is the bunching together of technical details so 
closely that the reader cannot read them at his normal 
reading rate. The writer makes little attempt to articu- 
late the details; he merely lays them down side by side. 
The impression on the reader is that he is being fed too 
rich a diet of technical information. 

The common forms of density in technical writing are 
(1) adjective strings, (2) stuffed paragraphs, (3) un- 
explained series, and (4) lumpy paragraphs. 

Adjective Strings 

In this form there is a high concentration of technical 
terms in front of a noun. The terms are either normal 
adjectives or adjectives made from nouns. Two examples 
from DEC software manuals appear below: 

1 . "The AFC1 1 is a flexible, high performance, 
multi-channel analog to digital (A/D) 

(Comment) Here flexible, high performance, 
multi-channel, and analog to digital are all 
adjectives qualifying converter. They com- 
press a wealth of technical data into a small 
space. For example, take flexible and high 
performance. A paragraph or two could be 
written on each one. But as they now stand, 
the writer hasn't bothered to define and 
analyze them for the reader. They are 
merely part of a cluster of details before a 
noun. And the reader is left with the job of 
interpreting them. 

2. "Because of the interactive, conversational, 
rapid-response nature of timesharing, a wide 
range of tasks — from solving simple mathe- 
matical problems to implementing complete 
and complex information gathering and 
processing networks — can be performed by 
the user." 


(Comment) This sentence contains two 
instances of piling up. What heightens the 
pain of the first one is that the adjectives are 
piled up before an abstract noun. Even so, 
the second one is more burdensome to the 
reader because there is a very real question 
as to the meaning of "implementing com- 
plete and complex information gathering 
and processing networks." If the reader 
expects to learn something from this sen- 
tence, he'll be disappointed. 

Happily, this kind of density is fast disappearing from 
DEC software documentation. 

Stuffed Paragraphs 

In this kind of density the paragraph is packed to the 
bursting point with technical details. For the most part, 
they are unanalyzed and unconnected. One thing all 
such paragraphs have in common is: They cannot be 
read and digested at a normal reading speed. Rather, 
they have to be poured over laboriously, at pretty much 
the speed used to decipher code or to locate an item in a 
catalog of television parts. 

Here are four stuffed paragraphs from our software 


1. "The DC71 remote batch station consists of 
a PDP-8/I processor, an operator Teleypte, a 
card reader, a line printer, and a synchro- 
nous interface. The DC71 connects to the 
DS10 or the DC75 via a full-duplex synchro- 
nous communications link. The remote 
batch terminal can be either a DC71A or 
DC71B terminal. The DC71A is configured 
with a 132-column line printer with a 
64-character set. The DC71B is configured 
with a 96-character set line printer. The 
DC71D Teletype Concentrator package 
includes eight lines for connecting to the 
DC71A or DC71B. Another eight lines can 
be added by connecting the DC71E to the 
DC71D. Terminals can be Teletypes, VT06 


or VT05 display terminals, or other Tele- 
type-compatible terminal interfaces, at 
speeds up to 2400 baud. 1 " 

(Comment) Perhaps the facts in this para- 
graph would look better in a vertical list — 
because that is exactly what the paragraph 
is: A list of very thinly related technical 
facts. A long sequence of such paragraphs 
would be deadening to the reader. In- 
evitably, they would cause the breakdown of 

The writer should do two things. First, he 
should ask whether all these technical facts 
are necessary. That is, does the reader of this 
manual need to know them all? If the 
answer is yes, then he should break up the 
paragraph into at least three smaller ones. 
Thus, he could consider the remote batch 
terminal, the line printer, and the other 
terminals as topics for separate paragraphs. 

"Single (DDI) or dual (DSP, DOV) displays 
are provided along with a set of display 
control commands. Two small vertical lines 
known as fixed cursors (DCU) and two 
bright crosses known as free cursors (DFR) 
can be displayed. Through these free cursors, 
it is possible to draw a straight line (DLI), 
which in most cases represents a base line. 
The display may also be expanded (DEX) 
from the data that is between the two fixed 
cursors. The total number of data points 
that are displayed may be increased (DIN), 
decreased (DDE), raised (DRA), or lowered 
(DLO), by a factor of two. The number of 
data points may also be changed to some 
arbitrary value (DPO) and viewed via a 
window (DWT). The table delta (DT A), used 
in selecting data points for display, can be 
varied. The ability to define the zero-data 
position of a display (BDD), and show the 
data expanded to the scope limits (DNO) is 
provided. The buffer identification name 
(DID), in conjunction with the free (DFV) 

teletype is a registered trademark of the Teletype Corporation. 


and fixed (DCV) cursor values, may be 
displayed. It is essential that a display be 
active in order that all display control 
commands work efficiently." 

(Comment) What an awesome pile of tech- 
nical data to squeeze into a single paragraph! 
Again, this is a mere catalog of details with 
absolutely no attempt to relate or interpret 
them. This information would be far more 
effective in a chart or table. Notice that it is 
impossible to insert logical connecting words 
or transitional expressions in a paragraph 
like this. In fact, the inability to put in such 
words and expressions is a mark of the 
stuffed paragraph. 

3. "Random access files, unlike sequential 
access files, do not distinguish between read 
mode and write mode. The user can read or 
write any item in a random access file at any 
time by first setting a pointer to that item. A 
random access file contains either string data 
or numeric data, but not both. Each data 
item in a random access file takes up the 
same amount of storage space, called a 
record, on the disk. BASIC must know the 
record size for the random access file in 
order to correctly move the pointer for that 
file from one data item to another. The 
record size for a random access numeric file 
is set by BASIC because the storage space 
required for a number in such a file is always 
the same. The storage space required for a 
string, however, is dependent upon the 
number of characters in the string. Thus, for 
a random access string file the user must 
specify the number of characters in the 
longest string in the file so that BASIC can 
set the record size accordingly. This specifi- 
cation takes place when the file is assigned 
to a channel. Refer to the description of the 
FILES and FILE statements in Section 10.2. 
When creating a new random access string 
file, if the user specifies too few characters 
an error message is issued when a string too 
long to fit into a record is written. If too 


many characters are specified for a record, 
the strings will always fit, but space will be 
wasted on the disk. When he is dealing with 
an existing file, the user does not have to 
specify a record size. If he does specify a 
record size for an existing file, the record 
size must match that with which the file was 

(Comment) Part of the difficulty with this 
paragraph is its size (301 words). But the big 
trouble is that its wealth of technical facts 
taxes the reader's memory. Too, there, is 
little unity and practically no coherence. 

The writer should try breaking the paragraph 
into three parts, as the first step in im- 
proving it. One paragraph could be devoted 
to read and write mode, another to records, 
and the last to strings. In addition, the writer 
should show explicitly the relationships 
among the facts so that the reader is helped 
through the text. That is, the writer should 
be liberal in the use of connectives and 
transitional expressions. 

"Scheduling may be forced before the time 
slice has expired if the currently running job 
reaches a point at which it cannot imme- 
diately continue. Whenever an operating 
system routine discovers that it cannot 
complete a function requested by the job 
(e.g., it is waiting for I/O to complete or the 
job needs a device which it currently does 
not have), it calls the scheduler so that 
another job can be selected to run. The job 
that was stopped is then requeued and is 
scheduled to be run when the function it 
requested can be completed. For example: 
when the currently running job begins input 
from a DEC tape, it is placed into the I/O 
wait queue, and the input is begun. A second 
job is scheduled to run while the input of 
the first job proceeds. If the second job then 
decides to access a DECtape, it is stopped 
because the DECtape control is busy, and it 
is placed in the queue for jobs waiting to 
access the DECtape control. A third job is 


set to run. The input operation of the first 
job finishes, freeing the DECtape control for 
the second job. The I/O operation of the 
second job is initiated, and the job is 
transferred from the device wait queue to 
the I/O wait queue. The first job is trans- 
ferred from the I/O wait queue to the 
highest priority run queue. This permits the 
first job to preempt the running of the third 
job. When the time slice of the first job 
becomes zero, it is moved into the second 
run queue, and the third job runs again until 
the second job completes its I/O 

(Comment) Once again you're dealing with a 
paragraph whose length (276 words) adds 
nothing to the ease of comprehension of the 
content. And once again, breaking it up is 
the best way to improve it. Certainly it can, 
with benefit, be broken in two at the 
example. Thereafter, the example itself must 
undergo some drastic ameliorative surgery. 
For as it now stands the tortuous rigmarole 
of the example serves more to obscure than 
to clarify. 

Unexplained Series 

A series of phrases or clauses bunched together too 
compactly is another kind of density. Here the writer 
crowds three or more advantages or functions into one 
sentence without showing the what or why of them. The 
reader, in likelihood, will skim through them without 
grasping their full significance. 

Two examples from a DEC software manual highlight 
this fault. 

1 . "The Real-Time Clock offers the user of the 
Lab Peripheral System several methods for 
accurately measuring and counting intervals 
or events. It can be used to synchronize the 
central processor to external events, count 
external events, measure intervals of time 
between events, or provide interrupts at 
programmable intervals. It can also be used 
to start the analog-to-digital converter by 


means of the overflow from the clock 
counter or by firing a Schmitt trigger. Many 
of these operations can be performed 

(Comment) The second sentence about how 
the real-time clock can be used is too dense 
and general. The writer should have devoted 
a sentence or two to each function. (The 
third sentence should have been expanded in 
the same way. And all the needless mystery 
can be swept out of the last sentence by 
telling which operations can be performed 

2. "Lab Applications-1 1 modules are available 
to perform operator console interaction, 
data acquisition, data editing, Fast Fourier 
transformation, output printing, and dis- 
playing. Lab Applications-1 1 allows many 
varieties of these functions. The library of 
modules will be enhanced as time goes on, 
and as application needs are defined, more 
and more of the requirements for laboratory 
computing will be supplied by DIGITAL." 

(Comment) The first two sentences embody 
a dense, but general, list. If they are in- 
tended to inform, they should be explained. 
If they are not intended to inform, they 
should be eliminated. 

(At the very least, the writer should expand 
on the statement about "many variations of 
these functions." The sentence is opaque as 
it stands.) 

Lumpy Paragraphs 

The term "lumpy" is applied to paragraphs containing 
an uneven distribution of technical detail. Some sen- 
tences in a lumpy paragraph are rich in technical 
content, others contain little or no technical matter. The 
net result of page after page of such paragraphs is to 
keep the reader continually off balance. 

The following examples from one of our software 
manuals show that the technical "lump" can appear 
anywhere in the paragraph. 


"The 1040 is the smallest of the five 
systems. The typical configuration of this 
system has a KA10 central processor, 32 to 
64K high-speed ME 10 core memories, the 
RP02G disk system with up to two disk 
packs, the TM10G magnetic tape system 
with up to two drives, and low-speed periph- 
eral equipment including a CR10F card 
reader, an LP10A line printer, and local 
DC 10 lines. This is an excellent system for 
the scientific research lab where multiple 
real-time tasks and general computing are 
required, and also for small colleges where 
there is a need for handling administrative, 
student, and faculty workloads simul- 
taneously. The system is easily expandable 
with most equipment on the DECsystem-10 
Equipment List." 

(Comment) In this instance the lump 
appears at the beginning. Sentence 2 con- 
tains all the technical fare of the paragraph. 
And, as you can see, the fare is too rich for 
comfortable digestion. Incidentally, one bad 
effect of having the knot of technical data 
appear at the beginning is that it obscures 
the main idea of the paragraph. (Remember 
that the beginning is the usual location of 
the topic sentence in technical writing.) 

"The 1077 is the dual-processor 1070 with 
fast execution of computing loads because 
of the second KI10 central processor. In 
addition, this system typically has 128K 
(640K bytes) of core memory, 690K words 
(4.1 million characters) of RM10G drum 
storage, a RP03G disk system with four disk 
drives for a total of 41.6 million words 
(249.6 million characters) of storage, a 
TU40G magnetic tape system with four 
120KC drives, a 1000 line-per-minute LP 10C 
line printer, a 1200 character-per-minutc 
CR10E card reader, and a DC 10 or DC68A 
communication system capable of 128 lines. 
In expanding to the 1077 from a smaller 
system, the user notices increased computing 
power, but he does not need to change his 


programs or learn a new command language 
or operating system." 

(Comment) In this three-sentence paragraph, 
the lump is located in the middle - in 
Sentence 2. And in its effect on communica- 
tion the lump is truly malignant. Look at its 
size. It consists of roughly 9 of the 15 lines 
of the paragraph. And it encompasses some 
13 ideas. A sprawling, technically rich knot 
of varied data, it serves to interrupt the 
natural flow of ideas from sentence 1 to 
sentence 3. 

"The 1055 is the dual processor equivalent 
of DECsystem-1050 with fast execution of 
compute-bound jobs because of the addition 
of the second processor. This system has two 
parallel KA10 processors connected with 
one operating system in order to double the 
computing power of the 1050 and at the 
same time to maintain the same interface 
between the user and the computing system. 
This approach of co-equal processors gives 
the user increased computing capacity when 
processing power is in heavy demand under 
multi-task loads. In addition to the two 
KA10 processors, the typical 1055 has 80K 
of ME 10 core memories, with one MX 10 
memory port multiplexer, one RM10G drum 
system, one RP03G disk system with up to 
eight disk packs, one TU40G, 120KC mag- 
netic tape system, one CR10 card reader, the 
LP 10C line printer, and 32 local lines, either 
a DC 10 system or a DC68A system." 

(Comment) Appearing in this instance at the 
end, the lump gives the whole paragraph a 
kind of foot-heavy look. It's like putting size 
1 8 shoes on a person who is five-seven. And 
setting aside any further comments on 
appearance, we can say that this lump is as 
much of a drag on communication as the 
lumps in Examples 1 and 2. 



The overall effect of density is to greatly reduce your 
chance of communicating with the reader. For one 
thing, a solid chunk of technical matter is likely to repel 
him because he finds it dull and unnatural. It disrupts his 
natural reading pace. For another, density interrupts the 
natural flow of ideas from sentence to sentence and 
paragraph to paragraph. Thus, he fails to get a sense of 
movement, or continuity, to carry him through the 
writing. In a word, he will probably resent dense writing 
because he had to plow through it. Finally, a dense 
presentation is hard for him to interpret. Hence, he is 
likely to go elsewhere for his information. Bearing in 
mind George Klare's principle of least effort, you know 
he won't work hard when he learns of an easier way to 
get the information. This is especially so if he figures 
that you should have done the interpreting for him. 

So in the interest of effective communication, examine 
your writing for the kinds of density discussed in this 
chapter. Read any questionable passages aloud to deter- 
mine whether they sound natural. If they turn out to be 
technical tongue twisters, think about density. And then 
set about revising them. Too, revise any passage in which 
you find it hard to insert connectives and transitional 
phrases. Such a test should again get you to think 

Once density is diagnosed, what's the cure? For a start, 
use the suggestions sprinkled throughout this chapter: 

1. Break up dense sentences and paragraphs. 
One spinoff benefit of such action is that 
you have to insert words showing relation- 
ships. Thus, you will automatically be inter- 
preting for the reader. 

2. Eliminate all unnecessary technical matter. 
Ask yourself: Does my reader need to know 
this? Consider substituting a familiar word 
for a technical term whenever possible. (But 
be cautious here. This advice must be used 
sparingly and with discretion.) 

3. Break up the lumps in lumpy paragraphs. 
Distribute the technical details among the 
various sentences. 


4. Use at least one sentence to explain each 
item in a dense series of features, functions, 
advantages, benefits, etc. 

5. Consider using tables and lists instead of 
paragraphs to present large clumps of tech- 
nical details. 

Perhaps, as part of the cure, you should ponder Thomas 
Johnson's law of density to see if you can use it as a 
guiding principle: "A reader's ability to understand a 
paragraph is inversely proportional to the number of 
technical terms that are present." 





". . . There's a glory for you!" 

"I don't know what you mean by 'glory,' " Alice said. 

Humpty Dumpty smiled contemptuously. "Of course 
you don't - till I tell you. I meant 'there's a nice 
knockdown argument for you!' " 

"But 'glory' doesn't mean *a nice knock-down argu- 
ment,' " Alice objected. 

"When / use a word," Humpty Dumpty said, in rather a 
scornful tone, "it means just what I choose it to mean — 
neither more or less." 

GLASS, Chapter 6) 

Humpty Dumpty is not just a product of Lewis Carroll's 
lively imagination. He actually exists. In fact, he 
abounds in the real world of words. He writes syndicated 
newspaper columns and learned articles for journals and 
magazines. His papers and reports appear in business and 
in science, in law and in medicine — indeed in every 
trade and in every profession. Sometimes he even stoops 
to teach English at a high school or college. But 
wherever he is, he always shows himself as the arch- 
enemy of communication. 

This chapter deals with three classes of words the 
Humpty Dumpty writer uses to hinder communication: 
(1) jargon, (2) big words, and (3) deadwood. 


The meaning of the term jargon has changed over the 
years. So nowadays jargon can be good or bad. 


In its good sense, it refers to the specialized technical 
language of a trade or profession. It is, in effect, a club 
language. Thus, medicine has a jargon, law has a jargon, 
baseball has a jargon. And the computer industry has its 
jargon. Some familiar examples of good computer jargon 
are debug, program, loop, core memory, subroutine, bit, 
block, branch, flag, checksum, software, flowchart, and 
tag. Being clear and concise in meaning, such words are 
excellent tools of communication. Writer and reader 
alike, as long as they belong to the same club, know 
what the words mean. 

However, when such words are used in communication 
with nonmembers — with readers who are not privy to 
their specialized meaning — then good jargon becomes 
bad jargon. 

For, in its derogatory sense, jargon refers to terms and 
expressions that are unclear either (1) because the 
audience is unfamiliar with them or (2) because writers 
attach new, private meanings to them. This second 
category is designed to impress the reader — to make 
him think the author is smart because he is "in." Such 
words may once have been the good jargon of a trade or 
profession. But they have since been robbed of their 
precise meaning. Now they are used by vague-thinking 
writers to hide the vagueness of their thinking. Some 
examples of computer terms now labeled bad jargon 
appear below: 

capability maximum support 

component operation 

compute-bound optimization 

computing power processing power 

concepts program parameter input 

data values reliability 

environment sophistication 

facility stand-alone system 

features system 

high-level language system resources 
high-speed performance technique 

implementation utilization 

Such words dare the reader to find out what they mean. 
But worst fault of all, they convey foggy ideas. On this 


ground they should be shunned as poor communication 

Another example of bad jargon (i.e., foggy words) is the 
too easily coined -ize words: Certainly we have a huge 
number of good -ize words: pasteurize, sterilize, magne- 
tize, oxidize, galvanize, anesthetize, phosphorize, etc. 
The bulk of these are precise scientific terms. But in a 
way that is unfortunate, for it leads readers to believe 
that the new -ize words — coined willy-nilly — are equally 
meaningful. Words like moisturize, parameterize, and 
vietnamize convey no precise idea at all. Another good 
example of this kind of jargon is the word finalize. It 
appeared in the following sentence in a recent DEC 
memo: "We intend to finalize the project plan Thursday 
afternoon." Does this mean 

1 . They intend to review the plan for the final 

2. They intend to insert all input from previous 

3. They intend to decide on exactly what will 
go into the plan and what will be excluded? 

4. They intend to have the plan signed by the 
people who must approve it? 

5. They intend to have it typed and ready to 
submit to the people who must approve it? 

6. They intend to submit it for final review by 
the people who must approve it? 

Most of the new -ize words leave the reader in this kind 
of quandary regarding their meaning. Thus it is that 
jargon works its harm on communication. Equally bad 
are the -wise words. Again, the suffix was formerly used 
with precision — to denote manner, direction, or 
position. Thus we have serviceable terms like clockwise, 
lengthwise, widthwise, and counterclockwise. But largely 
through the efforts of the commercial writers -wise came 
to mean with reference to. And readers are flooded with 
the likes of taxwise, pricewise, performancewise (which 
has been applied to everything from the stock market to 
partners in the sex act), saleswise, capitatwise, success- 
wise, weatherwise, profitwise, expensewise, loanwise, 
and even bottom-linewise. The computer industry con- 
tributed programwise, codewise, logicwise, bitwise, 
operationwise, recordwise, processingwise, fobwise, 
devicewise, and implementationwise. Given enough time, 


we might even come up with computing-powerwise or 

And finally there is a whole host of miscellaneous 
fog-words we likewise call jargon. For example, what 
makes end product and end result any better than 
product and result*! Only the autocratic say-so of the 
Humpty Dumpty writer. And the word objective*] Is it 
different from goal*! Hardly. It is just a high-sounding 
piece of jargon borrowed, perhaps, from the military. 
But when Humpty Dumpty writes of "achieving our 
goals and objectives," he gives it a private meaning he 
doesn't deign to pass on to the reader. And what about 
the word area! Some people write: "He will be working 
in the networks area." Or, "He will be in charge of all 
software in the communications area." Do those sen- 
tences say anything more than the following? 

1 . He will be working on networks. 

2. He will be in charge of all communications 

The Humpty Dumpty writer would have you believe 
they do. But don't be duped. Area here is just an empty 
four-letter word. Humpty Dumpty uses it because he has 
no interest in communicating. 


The modern Humpty Dumpty is adept at using big 
words — those of three syllables or more — to cloud his 
meaning. And he often uses them to impress the reader, 
rather than to inform him. Indeed, his motto seems to 
be: Never use a small word when a big one is at hand. 

The danger of big words lies in their many meanings. 
When readers cannot readily detect the writer's meaning, 
they are likely to pick a meaning at random. Thus, 
communication is left largely to chance. And as wc said 
in Chapter 1 , if the technical writer does not communi- 
cate, he is a failure. 

Typical of the big words that tend to mislead are those 
on this list culled from the first three pages of a DEC 
software manual. We are not saying they should be 
eliminated. Rather, we say they can impede communica- 
tion because they are more likely to be misunderstood 
than the smaller words you can put in their place. 


Big Word 

Small Substitute 








































greatest, highest 

















But combinations of big words create the densest fog of 
all. Then the reader is hit with clusters of meanings. And 
there is no telling what message he goes away with. 
Moreover, combinations of big words usually add up to 
long sentences. And long sentences, as the experts assert, 
are the greatest barrier to efficient communication. 

Three examples from DEC software documentation 
show how combinations of big words make the meaning 
unclear. Too, they show how word bigness and sentence 
length go hand in hand. 

1. (Example) "The KI10 processor provides 
measures for handling arithmetic overflow 
and underflow conditions, pushdown list 


overflow conditions, and page failure con- 
ditions directly by the execution of pro- 
grammed trap instructions instead of re- 
sorting to a program interrupt system." (36 

(Comment) Measures, three uses of condi- 
tions, and resorting to spread a meaning- 
concealing haze over the main idea in this 

2. (Example) "With the increased memory size, 
the high performance peripheral systems, 
and the large file system, the 1070 is 
configured for maximum support of remote 
batch capabilities through the synchronous 
communication equipment." (31 words) 

(Comment) Again, such mouth-filling terms 
as high performance, configured, maximum 
support, and capabilities serve to hide the 
writer's thoughts. 

3. (Example) "This approach of co-equal proc- 
essors gives the user increased computing 
capacity when processing power is in heavy 
demand under multi-task loads." (21 words) 

(Comment) Although this sentence is 
shorter, it has more heavy, multi-meaning 
words than the previous two. Approach, 
increased computing capacity, processing 
power, and multi-task loads protect the 
writer's meaning from the prying eyes of the 
reader. And the preposition under, though 
small, does nothing at all for communication 
in this sentence. 

So the advice in regard to vocabulary is to stick to short 
words whenever they fit. Every time you use a big word 
you sacrifice some of the meaning in your sentence. As 
E.B. Coleman and C.R. Miller learned from their 
research, short words are more efficient in conveying 


Finally, there is a class of foggy words known as 
deadwood. Because such words add nothing to the 


meaning of a passage, they are empty and lifeless. All 
they do is occupy space - making long sentences still 
longer and serving mostly to detract from the meaning 
of other words. A passage loaded with deadwood leaves 
the reader with the impression that he, like Shakes- 
peare's Hamlet, is reading "words, words, words." 

From a narrow point of view, deadwood is the needless 
repetition of ideas. Usually such repetition occurs 
because the writer is not paying close attention to what 
he is saying. He says "green in color," for example, or 
"four in number." But he knows full well that green has 
to be a color, and four can be nothing but a number. 
Other familiar examples are "round in shape," "the 
month of February," "five miles in distance," and "the 
year 1620." Like dead branches on a tree, these 
repetitions (in shape, the month of, in distance, and the 
year) are not essential and must be pruned. 

In larger perspective, deadwood refers to (1) long expres- 
sions that a single word can replace and (2) single words 
that add no meaning to a passage. As an example of an 
overlong expression, consider "by the use of." Three of 
the words are freeloaders doing absolutely nothing to 
carry the burden of meaning. Eliminate them and you're 
left with "by" as the effective substitute for the entire 
phrase. Similarly, other sentence-lengthening phrases can 
be reduced to a solitary word: 


Single-Word Repl 

at the rate of 


by means of 


for a period of 


for the purpose of 


in an area where 


in an effort to 


in order to 


in such a manner as to 


in terms of 


involves the use of 


is designed to be 


it is clear that 


it is evident that 


with the aid of 



On the other hand, an excellent example of a single 
word that often does nothing to further the meaning of 
a sentence is the word nature. It is, for instance, needless 
in the following context: 

The specifications are highly technical in nature. 

Thus, with the dead wood pruned the sentence becomes 

The specifications are highly technical. 

Notice that the sentence becomes smaller but the 
meaning remains unchanged. Invariably, this is what 
happens when empty words like nature are deleted. 
Similar examples follow: 


(Example) "After the group of modifications have 
been added to the file, RUNOFF produces a new 
copy of the file which is properly paged and 

(Revision) After the group of modifications have 
been added to the file, RUNOFF produces a new 
copy of the file which is paged and formatted. 

(Comment) Here, we can assume that paging and 
formatting are going to be done properly - 
especially when we're touting our own software. 


(Original) "TECO manipulates data within the 
editing buffer in response to suitable commands 
from the user." 

(Comment) Delete suitable. In a manual that 
describes all TECO commands, the notion of 
suitability does not have to be spelled out. The 
reader knows he won't get anywhere with the 
software unless he uses a suitable (i.e., legitimate) 

(Original) "The listing can be directed to any 
suitable output device: line printer, teletype, 
DECtape or disk." 


(Revision) The listing can be directed to an output 
device: line printer, teletype, DECtape or disk. 

(Comment) Labeling the devices as suitable is 
needless when they are listed. 


(Original) "The DECsystem-10 is more than a 
processor and its associated peripheral devices." 

(Comment) Delete associated. The word its shows 
the connection between processor and peripheral 


(Original) "This software allows him to define 
the hardware configuration of his particular 

(Comment) Delete particular. As in the preceding 
example, the possessive (his) serves to designate 
the installation. Particular just repeats - 


(Original) "Existing programs and data files can be 
modified directly with BASIC instead of with a 
system editor by adding or deleting lines, by 
renaming the file or by resequencing the line 

(Comment) Delete existing. Since modifying non- 
existing programs and data files would appear to 
be an impossibility, the use of existing is needless. 

However, when there is a kind of contrast between 
the existing and the nonexisting, the use of the 
word existing is valid. For example, the following 
usage from the same manual is all right: Com- 
mands to LINED allow the user to create a new 
file or edit an existing file. 


(Original) "This program determines by appro- 
priate dialogue with the user who he is, whether or 
not he is currently authorized to use the system, 


and if so, establishes the user's initial profile, 
informs him of any messages of the day, and 
reports any errors detected in his disk files." 

(Comment) Delete appropriate. Of course the 
dialogue is appropriate -j- for two reasons. First, 
the program cannot make an accurate determina- 
tion unless the dialogue is appropriate. And, 
secondly, appropriateness is taken for granted - 
hasn't this software been tested? 


(Original) "The project plan does not consider the 
prohibitive cost factor." 

(Revision) The project plan does not consider the 
prohibitive cost. 

(Original) "Field data proves conclusively that this 
monitor has a high reliability factor." 

(Revision) Field data proves conclusively that this 
monitor has high reliability. 


(Original) "For the last two years we have been 
actively supporting two monitors." 

(Comment) Delete actively. Unless of course we 
passively support other monitors. 


(Original) "There are three major components of 
the computing system: the actual machine, or 
hardware; the operating system, or monitor; and 
the languages and utilities, or nonresident 

(Comment) Delete actual. Machine and hardware, 
as everyone knows, exist in the actual world. So 
the emphasis provided by actual is not needed. 
Besides, aren't operating system, languages, and 
utilities also actual? Moreover, nowhere in this 
sentence is there any contrast with something not 



Words are the basic building blocks of your writing. And 
how you pick and choose among them determines your 
success in getting ideas from your head to the reader's. 
Choose words your reader understands, and you can 
build a vehicle to carry your ideas to him. Choose 
jargon, big words, and deadwood, however, and you 
build a barrier to stop your ideas in their tracks. For 
nothing can improve a passage whose words do not 
inform: not unity, not coherence, not the active voice — 
not any of the principles discussed earlier in this book. 
So do your reader a favor: pick building blocks that help 
him understand. Don't be another Humpty Dumpty. 




Adams, E.B. "A Comparison of Two Techniques of 
Presenting Information in a Public Information 
Bulletin." Dissertation for Doctor of Education 
Degree, University of Wyoming, 1972. 

Backus, M.G. "Conservation and Reading Comprehen- 
sion." Educational Resources Information Center, 
U.S. Office of Education (Document Number: ED 
094365), 1974. 

Bloomer, R.H. "Level of Abstraction as a Function of 
Modifier Load." Journal of Educational Research 
(Volume 52, pages 269-272), 1959. 

Borgh, Enola. "Transformations and Stylistic Options." 
Educational Resources Information Center, U.S. 
Office of Education (Document Number: ED 
078445), 1972. 

Bormuth, J.R. "Development of Standards of Reada- 
bility: Toward a Rational Criterion of Passage 
Performance." Educational Resources Information 
Center, U.S. Office of Education (Document 
Number: ED 054233), 1971. 

Botel, M., and Granowsky, A. "Syntactic Complexity 
Formula." Educational Resources Information 
Center, U.S. Office of Education (Document 
Number: ED 091749), 1972. 

Brewer, R.K. "The Effect of Syntactic Complexity on 
Readability." Dissertation for Doctor of Philo- 
sophy Degree, University of Wisconsin, 1972. 

Clapp, E.R. "Why the Devil Don't you Teach Freshman 
to Write?" Saturday Review (Volume 48, pages 
63-93), 1965. 

Coke, E.J. "Readability and Its Effects on Reading Rate, 
Subjective Judgments of Comprehensibility and 
Comprehension." Educational Resources Informa- 
tion Center, U.S. Office of Education (Document 
Number: ED 074152), 1973. 

Coleman, E.B. "The Comprehensibility of Several Gram- 
matical Transformations." Journal of Applied 
Psychology (Volume 48, pages 186-190), 1964. 


Coleman, E.B. "Learning of Prose Written in Four 
Grammatical Transformations." Journal of 
Applied Psychology (Volume 49, pages 332-341), 

Coleman, E.B. and Miller, C.R. "A Measure of Infor- 
mation Gained During Prose Learning." Reading 
Research Quarterly (Volume 3, pages 369-386), 

Coleman, E.B. and Miller, C.R. "A Set of Thirty-six 
Prose Passages Calibrated for Complexity." 
Journal of Verbal Learning and Verbal Belwvior 
(Volume 6, pages 851-854), 1967. 

Dansereau, D.F. "Analysis of the Reading Comprehen- 
sion Process: The Development and Utilization of 
an Assessment Technique and the Preliminary 
Exploration of Individual Differences in Perceiving 
Syntactic Complexities." Educational Resources 
Information Center, U.S. Office of Education 
(Document Number: ED 066726), 1972. 

Dittmer, A.E. "A Comparison of the Grammatical 
Structures of Professional, Senior High and Junior 
High Expository Writing." Dissertation for Doctor 
of Philosophy Degree, Wayne State University, 

Fagan, W.T. "The Relationship between Reading Dif- 
ficulty and the Number and Type of Sentence 
Transformations." Educational Resources Infor- 
mation Center, U.S. Office of Education (Docu- 
ment Number: ED 071051), 1971. 

Freedman, M. "Seven Sins of Technical Writing." 
College Composition and Communication, (Vol- 
ume 9, pages 10-16), 1958. 

Granowsky, A. "Background for a New Syntactic 
Complexity Formula." Educational Resources 
Information Center, U.S. Office of Education 
(Document Number: ED 083566), 1973. 

Harris, A.J. "Some New Developments in Readability." 
Educational Resources Information Center, U.S. 
Office of Education (Document Number: ED 
094344), 1974. 

Hebb, D.O. and Bindra, D. "Scientific Writing and the 
General Problem of Communication." American 
Psychologist (Volume 7, pages 569-573), 1952. 

Holmes, V.M. "Order of Main and Subordinate Clauses 
in Sentence Perception." Journal of Verbal 
Learning and Verbal Behavior, (Volume 12, pages 
285-293), 1973. 


Hutson, B.W. and Shub, J. "Developmental Study of 
Factors Involved in the Choice of Conjunctions." 
Educational Resources Information Center, U.S. 
Office of Education (Document Number: ED 
090575), 1974. 

Kowitz, G.T. "From ERIC Source Documents to 
Abstracts: A Problem in Readability." Educational 
Resources Information Center, U.S. Office of 
Education (Document Number: ED 086243), 

Lively, B.A. and Pressey, S.L. "A Method of Measuring 
the Vocabulary Burden of Textbooks." Educa- 
tional Administration and Supervision (Volume 9, 
pages 389-398), 1923. 

Lowe, H. "Solomon or Salami." Atlantic Monthly, 
(Volume 204, pages 128-131), 1959. 

Marcus, A. "Reading as Reasoning; Reading as Am- 
biguity: Understanding Sentence Structures." 
Educational Resources Information Center, U.S. 
Office of Education (Document Number: ED 
086950), 1971. 

McCullough, CM. "Preparation of Textbooks in the 
Mother Tongue, A Guide for Those Who Write and 
Those Who Evaluate Textbooks in Any Lan- 
guage." Educational Resources Information 
Center, U.S. Office of Education (Document 
Number: ED 082120), 1965. 

McGintitie, W.H. and Tretiak, R. "Sentence Depth 
Measures as Predictors of Reading Difficulty." 
Reading Research Quarterly (Volume 6, pages 
346-377), 1971. 

Meyer, B.J.F. "Structure of Prose: Identification and 
Effects on Memory." Educational Resources Infor- 
mation Center, U.S. Office of Education (Docu- 
ment Number: ED 076979), 1973. 

Meyer, B.J.F. and McConkie, G.W. "Effect of Position 
of Information in a Passage's Organizational Struc- 
ture on Recall." Educational Resources Informa- 
tion Center, U.S. Office of Education (Document 
Number: ED 090525), 1974. 

Mosberg, L. and Shima, F. "Comprehension of Con- 
nected Discourse." Educational Resources Infor- 
mation Center, U.S. Office of Education (Docu- 
ment Number: ED 036398), 1969. 

Palmer, W.S. "What Yolly, Willy, and Harriet Learned 
To Do - The Free Modifiers: A Fresh Mode of 
Teaching Composition." Educational Resources 


Information Center, U.S. Office of Education 
(Document Number: ED 059223), 1971. 

Powers, J.E. "The Effect of Children's Expectations and 
Word Associations Upon the Comprehension of 
Passive Sentences." Educational Resources Infor- 
mation Center, U.S. Office of Education (Docu- 
ment Number: ED 078931), 1973. 

Robertson, J.E. "Pupil Understanding of Connectives in 
Reading." Reading Research Quarterly (Volume 3, 
pages 387-417), 1968. 

Rothkopf, E.Z. and Kaplan, R. "Exploration of the 
Effect of Density and Specificity of Instructional 
Objectives on Learning from Text." Journal of 
Educational Psychology (Volume 63, pages 
295-302), 1972. 

Scott, R.I. "Using Korzybskfs Semantics to Teach 
English Composition." Educational Resources In- 
formation Center, U.S. Office of Education (Docu- 
ment Number: ED 040176), 1969. 

Slobin, D. "Grammatical Transformation and Sentence 
Comprehension in Childhood and Adulthood." 
Journal of Verbal Learning and Verbal Behavior 
(Volume 5, pages 219-227), 1966. 

Swift, M.H. "Clear Writing Means Clear Thinking." 
Harvard Business Review (Volume 51, pages 
59-62), 1973. 

Van Rooy, L. "Readability Studies and the Writer of 
Instructional Materials." Educational Resources 
Information Center, U.S. Office of Education 
(Document Number: ED 089245), 1973. 

Von Glasersfeld, E. "The Problem of Syntactic Com- 
plexity in Reading and Readability." Journal of 
Reading Behavior (Volume 3, pages 1-14), 1971. 

Walker, R.V. "The Technical Language Dimension: An 
Analysis of Contributing Factors." Dissertation for 
Doctor of Philosophy Degree, University of Cali- 
fornia, Irvine, 1972. 

Weaver, W.W., Kingston, A.J., Brickley, A.C., and White, 
W.F. "Information-Flow Difficulty in Relation to 
Reading Comprehension." Journal of Reading 
Behavior (Volume 1, pages 41-49), 1969. 

Williams, D.L. "The Effect of Rewritten Science Text- 
book Materials on the Reading Ability of Sixth- 
Grade Pupils." Dissertation for Doctor of Educa- 
tion Degree, University of Illinois, 1964. 

Williams, M. and Stevens, V.M.R. "Understanding Para- 
graph Structure." Journal of Reading (Volume 15, 
pages 513-516), 1972. 



(The books marked with an asterisk are recommended in 
their entirety to the new technical writer.). 

Barzun, J. and Graff, H.F. The Modern Researcher. 

Harcourt, Brace, and World, Inc., New York, 

*Black, M. Critical Thinking. Prentice-Hall, Inc., Engle- 

wood Cliffs, New Jersey, 1952. 
Blicq, R.S. Technically - Write! Prentice-Hall, Inc., 

Englewood Cliffs, New Jersey, 1972. 
*Brogan, J. A. Clear Technical Writing. McGraw-Hill 

Book Company, Inc., New York, 1973. 
*Casey, H. and Clark, M.T. Logic, A Practical Approach. 

Henry Regnery Company, Chicago, 1963. 
Crouch, W.G. and Zetler, R.L. A Guide to Technical 

Writing. The Ronald Press Company, New York, 

*Flesch, R. The Art of Plain Talk. Harper and Brothers, 

Publishers, New York, 1946. 
*Flesch, R. The Art of Readable Writing. Harper and 

Brothers, Publishers, New York, 1949. 
Flesch, R. A New Way to Better English. Harper and 

Brothers, Publishers, New York, 1958. 
Gilman, W. The Language of Science - A Guide to 

Effective Writing. Harcourt, Brace and World, Inc., 

New York, 1961. 
Gunning, R. The Technique of Clear Writing. McGraw- 
Hill Book Company, Inc., New York, 1952. 
*Hayakawa, S.I. Language in Thought and Action. 

Harcourt, Brace and World, Inc., New York, 1964. 
Hayakawa, S.I. ed. The Use and Misuse of Language. 

Fawcett Publications, Inc., Greenwich, Conn., 

Hicks, T. Successful Technical Writing. McGraw-Hill 

Book Company, Inc., New York, 1 959. 
Houp, K.W. and Pearsall, T.E. Reporting Technical 

Information. Benziger, Bruce and Glencoe, Inc., 

Beverly Hills, California, 1973. 
Institute in Technical and Industrial Communications: 

Fourth Annual Proceedings. ("Problems of Style 

and Semantics in Technical Writing," pages 16-20), 

University of Colorado, Fort Collins, Colorado, 

Institute in Technical and Industrial Communications: 

Eighth Annual Proceedings. Colorado State 

University, Fort Collins, Colorado, 1965. 


Institute in Technical and Industrial Communications: 
Eleventh Annual Proceedings. Colorado State 
University, Fort Collins, Colorado, 1968. 

*Johnson, T.P. Analytical Writing. Harper and Row, 
Publishers, Inc., New York, 1966. 

Klare, G.R. Measurement of Readability. Iowa State 
University Press, Ames, Iowa, 1963. 

Korzybski, A. Science and Sanity. The International 
Non-Aristotelian Library Publishing Company, 
(distributed by Institute of General Semantics, 
Lakeville, Connecticut) 1933. 

Menzel, D.H., Jones, H.M. Boyd, L.G. Writing a Tech- 
nical Paper. McGraw-Hill Book Company, Inc., 
New York, 1961. 

Morris, J.E. Principles of Scientific and Technical 
Writing. McGraw-Hill Book Company, Inc., New 
York, 1966. 

Mortenson, CD. and Sere no, K.K. Advances in Com- 
munication Research. Harper and Row, Publishers, 
Inc., New York, 1973. 

Nelson, J.R. Writing the Technical Report. McGraw-Hill 
Book Company, Inc., New York, 1952. 

O'Hayre, J. Gobbledygook Has Gotta Go. U.S. Govern- 
ment Printing Office, Washington, D.C., 1966. 

Peterson, M.S. Scientific Thinking and Scientific Writing. 
Reinhold Publishing Corporation, New York, 

Roberts, P. English Sentences. Harcourt, Brace, and 
World, Inc., New York, 1962. 

Roberts, P. Understanding English. Harper and Brothers, 
Publishers, New York, 1958. 

Schlesinger, I.M. Sentence Structure and the Reading 
Process. Moutone Company, N.V., Publishers, The 
Hague, 1968. 

*Shurter, R.L., Williamson, J.P., and Bruehl, W.G. 
Business Research and Report Writing, McGraw- 
Hill Book Company, Inc., New York, 1965. 

Sklare, A.B. The Technician Writes: A Guide to Basic 
Technical Writing. Boyd and Fraser Publishing 
Company, San Francisco, 1971. 

*Strang, B.M.H. Modern English Structure. William 
Clowes and Sons, Limited, London and Bcccles, 

Ulman, J.N., Jr. Technical Reporting. Henry Holt and 
Company, New York, 1952. 


UNESCO. Bibliography of Publications Designed to 
Raise the Standard of Scientific Literature. 
Unipub, New York, 1963. 

Weisman, H.M. Basic Technical Writing. Charles E. 
Merrill Books, Inc., Columbus, Ohio, 1962. 



Abstract words,l-7, 2-3, 2-8, 5-1, 5-2, 5-3, 54, 5-5, 
5-6, 5-7, 5-8, 5-9, 5-10, 5-11, 5-12, 5-13, 7-12, 

Abstract words as tools of thought, 5-6 

Abstract words vs. concrete words, 5-5 

Abstraction, high level of, 5-9 

Abstraction ladder, 5-2, 5-3, 5-4, 5-11, 5-12 

Abstraction process, 5-2 

Abstractions, detrimental, 5-8 

Active verbs, 5-7, 5-12 

Active voice, features of, 7-2 

Adverbial clauses, position of, 3-11 

Adverbs of time, 4-9 

Ambiguity, 5-7, 5-8, 5-10 

Ambiguity, prospective, 3-10, 6-2 

Artificiality, wall of, 2-4 

Ascending the abstraction ladder, 5-3 

Barzun, Jacques, 2-7, 4-7, 5-11,5-12 

Big words, 2-3, 9-1,94 

Boldface typography, 3-3 

Bormuth, J. R., 3-5, 3-6 

Boyd, L. G., 7-15 

Breach of contract, 4-5, 4-6 

Burden of communication, 1-5 

Catalog of facts, 4-10, 4-13 

Causc-and-effect relationships, 2-9, 4-12 

Chief concern of writer, 2-6 

Chronological order, 3-12 

Clark, H., 3-11 

Clause length, 3-11 

Clearness, 1-7, 2-2 

Coherence, 4-1, 4-6, 4-7, 4-8, 4-9, 8-7 

Coleman, E. B., 3-7, 5-7, 5-11, 5-12, 6-5,6-6 


art of, 1-5 

burden of, 1-5, 3-15 

effective, 14, 1-5, 1-6 

failure of, 3-7 

faulty, 1-5,5-8 

improvement of, 14 

poor, 1-7 

successful, 6-3 


INDEX (Cont.) 

Communication system, 4-13 
Communicator, 1-2, 3-5, 3-6, 3-11, 6-5, 6-6 

more effective, 4-13 

responsibilities of, 1-5 
Communis, 1-2 
Competence in thinking, 2-2 
Competence level in writing, 2-1, 2-2, 2-4, 2-5, 2-6, 

Competitors, manuals of, 2-6 
Complex, simplifying the, 1-6 
Complex sentence, 1-7, 6-1 
Complexity, perceptual, 3-13 
Complexity, temporary, 6-5 
Compound sentences, 6-10 
Comprehension, 3-6, 3-7, 3-11, 3-13 
Comprehension difficulty, 3-11 
Concrete examples, 5-12, 5-13 
Concreteness, 2-3 
Concrete reference, 5-4 
Concrete word, 5-5, 5-6 
Conditional clause, 3-10 
Conjunctions, simple, 4-9 
Conjunctions, subordinate, 4-9 
Connected ideas, 4-7 
Connectives, 4-1, 4-11, 4-12, 8-7, 8-12 
Connotation, 1-6 
Consistency, 4-2 
Continuity, 4-6,8-12 

Contract between writer and reader, 3-12, 4-2, 4-5,4-6 
Conversational style, 2-5 

Dangling modifiers, 7-6 
Dansereau, Donald, 3-8 
Darnell, E. K., 3-12 
Deadwood, 9-6 
Decoding the message, 1-6 
Denotation (literal meanings), 1-6 
Density, 2-11,8-1 

adjective strings, 8-3 

lumpy paragraphs, 8-9 

stuffed paragraphs, 8-4 

unexplained series, 8-8 
Descending the abstraction ladder, 5-3 
Directness, natural, 2-4 
Doer of the action, 7-2, 7-3, 7-13 


INDEX (Cont.) 

Effective coherence, 4-8 

Effective communication, 1-4 

Effectiveness, 2-2 

Emphasis, lack of, 4-11 

Equal ideas, 2-9 

Equality and subordination of ideas, 2-2 

Exactness, 2-2 

Exposition, 1-6,4-11,5-8 

Exposition, changes in logical order of, 3-12 

Facts, catalog of, 4-10, 4-13 
Facts, grammatically equal, 4-1 1 
Flesch, Rudolf, 3-8, 5-11, 5-13, 6-7 
Flow of thought in written material, 4-7 
Foggy words, 9-2, 9-3, 9-6 

big words, 9-5 


jargon, 9-1 
Freedman, Morris, 5-8, 7-1, 7-10 

General concrete words, 5-5 
Generality level, 5-3, 5 A 
Graff, Henry, 2-7, 4-7, 5-11, 5-12 
Grammar, 2-1 

reader's, 3-7 

writer's, 3-7 
Grammatical equality of facts, 4-1 1 
Granowsky, Alvin, 7-5 
Gulf between writers and readers, 2-3 
Gunning, Robert, 2-2, 3-8, 3-14, 4-3, 5-5, 5-8, 5-11, 

Gunning's warning, 2-2 

Hard words, 2-3 
Haystack paragraph, 4-10 
Hicks, Tyler, 3-4 
Holmes, V.M., 3-11, 3-13 
Houp, Kenneth, 6-7 
Humpty Dumpty writer, 9-4 


connected, 4-7 
equality of, 2-2, 2-9 
foggy, 9-2 
identity of, 4-7, 4-11 


INDEX (Cont.) 

kind and order of, 2-1, 2-2, 4-7 

natural flow of, 8-12 
Illiterate graduates, 3-6 
Image of reader, 3-1 
Impressing the reader, 1-2 
Increasing comprehension, 3-13 
Information, position in paragraph, 3-10 
Information processing, 34, 6-2 
Instructional manuals, 3-1 1 
Inverted sentences, 2-5 

Jones, H. M., 7-15 

Jargon, 9-1 

Johnson, Thomas, 2-3, 2-10, 4-4, 5-8, 5-11, 5-13, 7-4, 

7-10, 7-14, 8-13 
Joint trouble, 4-8 

Klare, George, 2-6, 3-1, 3-7, 3-8, 3-9, 3-14, 6-6, 8-12 
Korzybski, Alfred, 5-1 

Ladder of abstraction, 5-3,5-5, 5-6 y 5-9, 5-12 

Language, simpler, 3-2 

Language and reality, 5-1 

Language as a tool, 1-6 

Learning, motive for, 3-2 

Literacy level, 2-1, 2-4 

Literal meanings (denotation), 1-6 

Links, 3-13 

Logical connections, 4-11, 4-12 

Logical order, 3-12,3-13 

Logical shorthand, 5-6 

Long paragraphs, 3-14 

Long sentences, 3-8, 3-9, 6-10, 9-5 

Long words, 2-8,3-8,3-9 

Lowe, Helen R., 3-6, 3-8 

McMahon, L., 3-11 

Main idea of paragraph, 2-10 

Making abstract sentences more meaningful, 5-11 


instructional, 3-11 

tutorial, 3-11 

four forms of, 1-3 

implied, 1-6 

multiple, 1-4 


INDEX (Cont.) 

Meanings, 5-7, 5-8, 5-9, 9-3, 9-4 

Means of transmission, 1-3, 1-6 

Median number of years of schooling, 3-2 

Memory level, required, 6-3 

Menzel, D. H., 7-15 

Meyer, Bonnie, 3-10 

Michaelson, Herbert, 3-3, 5-8 

Miller, C. R., 3-7, 5-7, 5-11, 5-12, 6-6 

Morris, Jackson, 6-3, 6-5 

Mosberg, Ludwig, 4-1 

Motion, 4-1 

Noise, 1-3, 1-7 
origin of, 1-7 

Object complements, 3-13 

logical, 3-12, 3-13 

sequential, 3-12 

chronological, 3-12 
Order of ideas, 2-1, 3-13 
Opening paragraphs, 2-10 
Organization, 2-2, 4-1 

lack of, 2-3 

poor, 1-7 

tight, 4-1 

Pacing, fast, 8-2 

Pallid passives, 7-4, 7-11, 7-15 


haystack, 4-10 

length of, 8-8 

main idea of, 2-10 

opening, 2-10 

organization of, 4-2, 4-9 

per page, 3-14 

position of information in, 3-10 

structural, 3-14 

unorganized, 4-1, 4-13 
Passive in literature and technical writing, 7-14 
Passive voice, 1-7, 2-8, 7-2 

abstractness of, 7-12 

case against, 7-1,7-4 

case for, 7-12 

complexity of, 7-5 

concealment, 7-5 


INDEX (Cont.) 

deemphasis, 7-11 

misconstruction, 7-6 

negative, 7-5 

recognizing the, 7-1 

shifting to the active from, 7-1, 7-15 

wastefulness of, 7-10 

wordiness of, 7-10 
Pearsall, Thomas, 6-7 
Perceptual complexity, 3-13 
Phrase, choppy, 2-9 
Picturable words, 5-6, 5-9 
Position of adverbial clauses, 3-1 1 
Principle of least effort, 3-14, 8-12 
Profile of reader, 3-1 
Prospective ambiguity, 3-10, 6-2 
Punctuation, 2-1 
Purpose of exposition, 1-6 

Readability, 2-6, 3-1, 3-14, 6-6, 6-10 
Readability tests, 3-1, 6-6 
Reader, 3-1 

burden of interpretation, 4-1 1 

challenging the, 3-14 

comprehension, 4-5 

defrauded by writer, 4-12 

educational level of, 3-2 

energy of, 3-8, 3-9 

examples familiar to, 3-3 

eye movements of, 3-7 

focusing attention of, 4-2 

grade level of, 3-14 

image of, 3-1, 3-4 

inefficiency of, 3-4, 3-6, 3-8, 3-10, 3-11, 3-12, 
3-14, 3-15 

memory of, 3-9, 3-11, 6-3, 6-6, 8-7 

mental power of, 3-9 

motivation of, 3-2, 3-14 

needs of, 1-4, 1-6, 2-10, 3-1, 3-4, 8-5, 8-12 

normal reading pace of, 2-1 1 

poor, 3-5,3-6,3-7, 3-14 

processing information, 3-4 

publications depend on, 2-2 

remembering facts, 6-3 

specialized background, 3-2 

template of, 3-4 


INDEX (Cont.) 

terms familiar to, 3-3 

typical, 3-4 

voluntary, 3-2 

wants of, 2-3 
Reader and complex sentences, 3-6 
Reader orientation, 3-14 
Reader profile, 3-1 

administrator or technical person, 3-4 

attention to the material, 3-3 

intellectual level, 3-3 

interests, 3-3 

selecting a reader, 3-4 

subordinate in knowledge, 3-3 

superior in knowledge, 3-3 
Readership, 2-6, 3-5 
Reading errors, 3-6 

Reading level and grade completed, 3-2 
Reading of textbooks, 3-5 
Receiver, 1-3, 1-4, 1-6 
Receiver of the action, 7-3 
Reference, unclear, 2-7 
Relationship between words and the external world, 

Relationships in writing, 2-2 

cause-and-effect, 4-12 

implied logical, 4-11 

logical, 4-11 
Relevance, 2-2 
Required memory level, 6-3 
Revision, 2-5 

Schools better, 3-5 
Self-expression, 1-2, 1-6, 2-3 
Sender, the, 1-3, 1-5 

burdensome, 64 

complexity of, 1-7, 6-1, 6-10 

length of, 6-6 

long, 2-4 

loosely associated, 4-7 

short, 2-3, 2-5, 3-7, 3-14, 6-6, 6-7, 

tightly linked, 4-7 
Sentence structure, 3-9 
Sequential order, 3-12 
Shifts of topic, 4-5 


INDEX (Cont.) 

Shima, Fred, 4-1 

Short words, 2-5, 3-7, 3-14, 9-6 

Signals, 4-12 

Simplifying writing for the reader, 2-5 

Slobin, Dan, 7-5 

Smith, K., 3-11 

Source, the, 1-3, 1-4, 1-5 

Speaker, clear, 2-4 

Specific concrete words, 5-5, 5-6 

Spencer, Herbert, 3-9 

Struck, Herman, 7-1 

Structural cues in written material, 3-8 

Structural paragraphs, 3-14 

Structure, inverted, 3-9 

Structure of speech, 6-1 1 

Style, impersonal, 2-3 

Subject complements, 3-13 

Syntax, false cues of, 2-7 

Tables, quick-reference, 3-3 
Technical writer, 

function of, 1-2 

unique skill of, 1-2 

use of the term, 1-2 
Technical writing, 

essential function of, 2-12 

more readable, 2-3, 2-12 

six major ills of, 1-7 
Terms familiar to the reader, 3-3 
Testing and measurement, 2-1 
Thinking, complex, 2-4 
Tone of writing, 3-3 
Topic, shifts of, 4-5 
Topic of paragraph, 4-2 
Topic sentences, 3-10, 4-2 
Transformation grammar, 6-5 
Transitions, 3-13, 44, 4-7, 4-9, 8-6, 8-12 
Tutorial manuals, 3-11 

Unity, 4-1, 4-2, 4-3, 44, 8-7 
Unorganized paragraphs, 4-1, 4-13 
Usage, 2-1 

van Rooy, Lois, 3-11, 5-6 

Verbal communication system, 1-3, 3-1 


INDEX (Cont.) 

Verb-nominalization, 5-12 

active, 5-12, 7-1, 7-2 

occurrence of main, 3-13 

passive, 7-1, 7-2 
Voice in grammar, 7-1 
Voice in your writing, 2-5 
von Glasersfeld, Ernst, 3-10 

"Weakness at the joints", 4-7 

Word and referent, 5-2, 5-3 

Word order of the declarative sentence, 6-5 


abstract, 5-1,5-5 

as shorthand terms, 5-2 

big, 3-7, 9-1 

concrete, 5-5,5-6 

fuzzy, 5-13 

general, 5-2 

many-syllable, 3-7 

picturable, 5-6, 5-9 

recognition of, 3-9 

small, 3-14 

transitional, 4-4 

vague, empty and difficult, 1-7 

chief concern of, 2-5, 26 

ordered thinking of, 4-8 

purpose of, 2-6 

right attitude of, 2-5 

style of, 1-7 

wrong attitude of, 2-3 
Writer-reader contract, 3-12, 4-2, 4-5,4-6 

bad habits of, 2-4 

complex, 2-4 

confusion of thought in, 2-9 

four-step approach to, 2-5 

goal of, 2-4 

reading it aloud, 2-5 

tone of, 3-3 

unorganized, 4-10 
Writing to be understood, 3-15 
Writing to impress, 2-4 


Printed in U.S.A.